Tese FINAL_ThainaLessa.pdf
Documento PDF (3.9MB)

Exibir conteúdo Baixar

                    UNIVERSIDADE FEDERAL DE ALAGOAS
INSTITUTO DE CIÊNCIAS BIOLÓGICAS E DA SAÚDE
Programa de Pós-Graduação em Diversidade Biológica e Conservação nos
Trópicos

THAINÁ LESSA PONTES SILVA

LIDANDO COM A INCERTEZA E IGNORÂNCIA NOS DADOS DA BIODIVERSIDADE:
uma perspectiva taxonômica e espacial

MACEIÓ - ALAGOAS
Março/2024

THAINÁ LESSA PONTES SILVA

LIDANDO COM A INCERTEZA E IGNORÂNCIA NOS DADOS DA BIODIVERSIDADE:
uma perspectiva taxonômica e espacial

Tese apresentada ao Programa de PósGraduação em Diversidade Biológica e
Conservação nos Trópicos, Instituto de
Ciências
Biológicas
e
da
Saúde.
Universidade Federal de Alagoas, como
requisito para obtenção do título de Doutor
em CIÊNCIAS BIOLÓGICAS, área de
concentração
em
Conservação
da
Biodiversidade Tropical.
Orientador: Prof. Dr. Richard James Ladle

MACEIÓ - ALAGOAS
Março/2024

Catalogação na fonte
Universidade Federal de Alagoas
Biblioteca Central
Divisão de Tratamento Técnico
Bibliotecária: Girlaine da Silva Santos – CRB-4 – 1127
S586l Silva, Thainá Lessa Pontes
Lidando com a incerteza e ignorância nos dados da biodiversidade: uma
perspectiva taxonômica e espacial / Thainá Lessa Pontes Silva. – 2024.
104 f.: il.: color.
Orientador: Richard James Ladle.
Tese (Doutorado em Ciências Biológicas) - Universidade Federal de Alagoas.
Instituto de Ciências Biológicas e da Saúde. Programa de Pós-Graduação em
Diversidade Biológica e Conservação nos trópicos, Maceió, 2024.
Inclui bibliografias.

1. Biodiversidade. 2. Lacuna Linneana. 3. Taxonomia. 4. Lacuna Wallaceana.
5. Distribuição geográfica. 5. Biogeografia. I. Título.
CDU: 574

Agradecimentos
Eu sempre gostei de pensar que os resultados dos projetos que a gente
desenvolve, seja no TCC ou no doutorado, nos levam a contar uma história. Houve
momentos que eu achava que a história desse meu doutorado não fazia sentido, e a prof.
Ana Malhado, sempre otimista, dizia que ia dar certo, que perto do final essa história
ficava encaixada. Bom, no final realmente deu certo. Eu me sinto muito orgulhosa do
que foi feito até aqui, da história que vivi e contei nesses 4 anos.
Enfrentamos uma pandemia, vi meus orientadores atravessarem o oceano, sai da
casa da minha mãe, fui morar com meu companheiro, trabalhei muito, viajei muito, fui à
África e à Portugal 2x, passei por uma cirurgia, tive três sobrinhos lindos, e, é claro,
comecei a terapia. Entre inseguranças e felicidades. Entre frustrações e realizações.
Reconhecendo minhas qualidades e limitações. Consegui. Mas consegui porque tive
muito apoio e colaboração de gente querida.
Então gostaria de agradecer aos meus orientadores, Richard Ladle e Ana
Malhado, que tanto acreditaram, me apoiaram e me proporcionaram oportunidades
incríveis nesses últimos anos. Estamos juntos no Lacos21 há mais de 10 anos, eles
sempre me tirando da zona de conforto, mas eu não consigo dizer não.
Agradeço também ao meu grupinho SOS (Evelynne, Cacá, Carol e Ludmilla),
minhas colegas de doutorado e meus amigos de laboratório, que dividimos tantos
momentos, entre cafés e cervejas, nesse laboratório só tem pessoas/profissionais que
tanto admiro e me inspiro. Eu sempre digo que sou formada em Lacos21.
Agradeço aos colaboradores dos artigos, em especial a Ju Stropp e Joaquín
Hortal que estiveram comigo no primeiro capítulo; a Karol com K que segurou minha
mão e me deu forças no segundo capítulo; e a Fernanda e o Javí que tanto me ajudaram
e apoiaram no terceiro capítulo. Aprendi demais e me senti muito honrada em trabalhar
com cada um de vocês.
Agradeço a Capes e ao Tropibio pelo financiamento para que essa pesquisa fosse
realizada, apoiando com a bolsa de estudo, com as viagens e publicações dos artigos.
Agradeço a todos os membros de bancas que tive no decorrer deste processo.
Em especial a banca de defesa, Prof. Robson, Prof. Guilherme, Dra. Geiziane e Dra.
Filipa, que foi escolhida com muito carinho na certeza de que teríamos uma troca muito
respeitosa e engrandecedora.

Agradeço ao meu companheiro, Gaspar, que é meu Porto Seguro, meu ombro
amigo e meu fã nº 1, ele é a pessoa que mais me apoia e impulsiona, ainda por cima
dividiu o peso de um doutorado junto comigo.
Agradeço minha família, minha mãe, meu pai, padrasto, irmãos, tios, primos,
avó que mesmo sem entender muito bem o que eu faço, me apoiam e se orgulham de
mim. Minha mãe, Stella, sempre falou para estudar e não ficar atrás do balcão. Mainha,
acho que já tá bom de estudar, não é?
Agradeço aos meus sobrinhos (Gabriel, Caetano e Bernardo), e suas mães,
minha irmã gêmea Thais, minha prima-irmã Juliane e minha amiga-irmã Yalli. Foram
meu fôlego e ânimo nos momentos mais delicados. Cada sorrisinho deles, palavra de
apoio delas, foi muito importante para mim.
E por fim, e não menos especial, eu agradeço as minhas amigas e meus amigos,
sem citar nomes, pois tenho medo de esquecer de alguém. Mas se eu fechar os olhos, eu
me lembro de cada um que entendeu meu cansaço e minha ausência, quando estava
viajando ou quando neguei saídas para trabalhar, ou quando estava longe, quando
deveria estar perto. Obrigada pelo apoio, colo e motivação. Atravessei esse rio, que foi
uma vida, e estive muito bem acompanhada.

Resumo
Esta tese, composta por uma revisão da literatura e três capítulos em formato de
manuscrito, oferece uma valiosa contribuição para a compreensão dos desafios
relacionados à ignorância e incerteza nos dados taxonômicos e espaciais da
biodiversidade. A revisão da literatura traz uma visão geral acerca das lacunas Linneana
e Wallaceana. No primeiro capítulo, exploramos os múltiplos conceitos e métricas
associadas à definição de espécies, ressaltando a complexidade e as incertezas
taxonômicas resultantes. Enfatizamos a influência das mudanças taxonômicas nas
estimativas de riqueza de espécies, destacando a necessidade de incluir informações
sobre estas mudanças para melhorar a precisão das estimativas de biodiversidade. No
segundo capítulo, a pesquisa ainda se concentra na lacuna Linneana relacionada à
nomenclatura de aves globais, apresentando uma métrica de incerteza taxonômica, e
avaliando associações entre características biológicas e ecológicas das aves à incerteza.
O terceiro capítulo aborda a lacuna Wallaceana, avaliando as lacunas temporais e
espaciais no esforço amostral da biodiversidade na Namíbia, utilizando dados do GBIF,
e analisando qual a influência de variáveis sociogeográficas no esforço amostral.
Concluímos evidenciando os desafios enfrentados por cientistas e coletadores em várias
dimensões do conhecimento da biodiversidade, desde as definições de espécies à coleta
e compartilhamento de dados. Ressaltamos a importância de investir em ciência de base
para preencher as lacunas e garantir a ampla disseminação do conhecimento taxonômico
e espacial das espécies.
Palavras-chave: lacunas do conhecimento, biodiversidade, taxonomia, distribuição,
conservação.

Abstract
This thesis consists of a literature review and three chapters in manuscript format. We
offer a valuable contribution to understanding the challenges related to ignorance and
uncertainty in taxonomic and spatial biodiversity data. The literature review provides an
overview of the Linnean and Wallacean shortfalls. In the first chapter, we explore the
multiple concepts and metrics associated with species definition, highlighting the
complexity and taxonomic uncertainties. We emphasize the influence of taxonomic
changes on species richness estimates, pointing out the need to include information on
these changes to improve the precision of biodiversity estimates. In the second chapter,
the research still focuses on the Linnean shortfalls related to the nomenclature of global
birds, presenting a metric of taxonomic uncertainty, and evaluating associations
between biological and ecological characteristics of birds and taxonomic uncertainty.
The third chapter addresses the Wallacean shortfalls, assessing the temporal and spatial
gaps in the biodiversity sampling effort in Namibia, using GBIF data, and analyzing the
influence of sociogeographic variables on the sampling effort. We conclude this thesis
by arguing the challenges faced by scientists and collectors in several dimensions of
biodiversity knowledge, from species definitions to data collection and sharing. We
emphasize the importance of investing in basic science to fill the gaps and ensure the
broad dissemination of taxonomic and spatial knowledge of species.
Key-word: knowledge shortfalls, biodiversity, taxonomy, distribution, conservation.

Lista de Figuras
Revisão da Literatura
- Figura 1: Proporção de espécies atualmente aceitas em relação ao total de descrições
taxonômicas feitas a cada ano; As tonalidades das cores indicam o número de
descrições taxonômicas feitas a cada ano em escala logarítmica, sendo que as
tonalidades mais claras indicam maior número de descrições. Os painéis a) e b)
representam dados de duas famílias de plantas, Fagaceae e Solanaceae, enquanto os
painéis c) e d) o fazem para duas famílias de peixes, Cychlidae e Characidae. Fontes de
dados: A lista de nomes de táxons, status taxonômico e ano de publicação foi
recuperada de Govaerts (2022) para as famílias de plantas, e de Froese et al. (2022) para
as famílias de peixes. Créditos da imagem: silhuetas foram baixadas em
https://thenounproject.com; Fagaceae (carvalho de Eucalyp); Solanaceae (tomateiro de
Michael Zick Doherty); Cychlidea (Cichlid por Ametyst Studio); Characidae (piranha
de Agne Alesiute). Figura elaborada pela Dra. Juliana Stropp, compartilhada com
permissão. Os dados e o código R estão disponíveis em https://github.com/justropp... 23
- Figura 2: Mudança taxonômica em palmeiras amazônicas (Arecaceae; Palmae). O
painel (a) mostra a reclassificação taxonômica de 708 sinônimos em 148 nomes aceitos;
cada cor representa um nome de espécie. O painel (b) descreve detalhadamente a
ligação entre sinônimos heterotípicos e nome aceito em uma linha do tempo de
agrupamento taxonômico (lumping) para um desses 148 nomes aceitos, Attalea
butyracea; linhas coloridas horizontais marcam o ano de descrição de 18 sinônimos
heterotípicos e o ano de sinonimização, sendo que cada cor representa um sinônimo
atual; as linhas verticais indicam a contagem de nomes de espécies aceites num
determinado ano; linhas curvas representam o agrupamento de 18 sinônimos
heterotípicos em A. butyracea. Fontes de dados: para o painel (a) a lista de nomes
aceitos de palmeiras amazônicas foi extraída de (Cardoso et al., 2017; ter Steege et al.,
2019), enquanto a lista de sinônimos foi obtida de (Govaerts et al., 2022); e a lista de
sinônimos heterotípicos mostrada no painel (b) foi obtida de Henderson (2020). Figura
publicada
em
short
communication
no
Journal
of
Biogeography
(https://doi.org/10.1111/jbi.14463), compartilhada com permissão dos autores........... 25
Capítulo 1: How taxonomic change influences forecasts of the Linnean Shortfall
(and what we can do about it)
- Figure 1: Schematic representation of the impact of lumping and splitting on estimates
of the number of known and unknown species (Linnean Shortfall). The centre of the
graph represents taxonomic stability (no further splitting or lumping of species). A
scenario of more splitting than lumping leads to underestimates of the shortfall (right
lower quadrant, blue), while more lumping than splitting leads to overestimates of the
shortfall (left upper quadrant, yellow). For example, there was relatively more splitting
than lumping of mammals between 2005 and 2017 (figures from Burgin et al. 2018),
meaning pre-2005 estimates of the total number of mammal species (unknown +
known) were almost certainly underestimates. Size of animal silhouettes is proportional
to the number of valid species, with grey indicating pre-2005 and black post-2005…. 44

Capítulo 2: Do biological and ecological variables influence nomenclatural
uncertainty in birds?
- Figure 1: Bar chart representing the nomenclatural agreement of extant world bird lists
in comparison with a reference list (IOC 2023.1). Dark blue bars represent the number
of completely discordant species (genus and species different), medium blue represents
change in genus (medium blue) and light blue represents a change in species name. The
number next to the bird silhouette refers to the total number of bird species in each list.
The percentages in the graph have been rounded to the nearest whole number. For full
values, see Supplementary material 1…………………………………………………. 64
- Figure 2: Column graph representing the percentage of the number of species with
nomenclatural uncertainty in bird Orders. The table below is the total number of species
(green) and number of species with nomenclatural uncertainty (red) per Order. Boxplot
(orange) shows the distribution of nomenclatural uncertainty in the Orders.
Accipitriformes
(ACC),
Anseriformes
(ANS),
Bucerotiformes
(BUC),
Caprimulgiformes (CAP), Cariamiformes (CAR), Cathartiformes (CAT),
Charadriiformes (CHA), Ciconiiformes (CIC), Coliiformes (COL), Columbiformes
(CLB), Coraciiformes (COR), Cuculiformes (CUC), Eurypygiformes (EUR),
Falconiformes (FAL), Galliformes (GAL), Gaviiformes (GAV), Gruiformes (GRU),
Leptosomiformes (LEP), Mesitornithiformes (MES), Musophagiformes (MUS),
Opisthocomiformes (OPI), Otidiformes (OTI), Passeriformes (PAS), Pelecaniformes
(PEL), Phaethontiformes (PHA), Phoenicopteriformes (PHO), Piciformes (PIC),
Podicipediformes (POD), Procellariiformes (PRO), Psittaciformes (PSI), Pterocliformes
(PTE), Sphenisciformes (SPH), Strigiformes (STR), Struthioniformes (STT),
Suliformes (SUL), Trogoniformes (TRO)...................................................................... 65
Capítulo 3: Quantifying spatial ignorance in the effort to collect terrestrial fauna
in Namibia, Africa
- Figure 1: Map of Namibia highlighting socio-geographical variables used in our
analysis, for example, roads, protected areas, vegetation cover and density of people.. 81
- Figure 2: Historical progression of the number of occurrence records for Namibia's
biodiversity publicly available on GBIF platform (full dataset). Number of records of all
taxa (grey line) and separately, according to the fauna silhouette…………………….. 87
- Figure 3: Bar graphic indicates the number of records in full dataset (1783-2021) and
recent dataset (2000-2021). Radar chart illustrates the percentages of GBIF’s basis of
records using full dataset, with HO=Human Observation and PS=Preserved
Specimen……………………………………………………………………………… 88
- Figure 4: Spatial distribution of GBIF’s records and ignorance scores for Namibia’s
birds, mammals, reptiles, amphibians and insects. Maps were calculated from recent
dataset (2000-2021). Ignorance maps represent a gradient of ignorance scores - from
cells with high ignorance scores (purple tons) to cells with low ignorance scores (yellow
tons). Histograms represent the frequency of cells according to ignorance scores
gradient. Silhouettes refer to taxonomic groups………………………………………. 89

Lista de Tabelas
Revisão da Literatura
- Tabela 1: Principais abordagens para estimar a lacuna Linneana (modificado de
MORA et al., 2011)........................................................................................................ 19
Capítulo 2: Do biological and ecological variables influence nomenclatural
uncertainty in birds?
- Table 1: Examples of nomenclatural uncertainty among scientific names on world bird
lists…………………………………………………………………………………….. 61
- Table 2: Biological and ecological variables used to explain nomenclatural uncertainty
among bird species. The table provides a brief assumption of why variables were
included into the model and the source of data collected……………………………... 63
- Table 3: Significant results from the Beta regression model analysing the association
between nomenclatural uncertainty and biological and ecological variables of bird
species. Full results with non-significant associations are available in Supplementary
material 3……………………………………………………………………………… 66
Capítulo 3: Quantifying spatial ignorance in the effort to collect terrestrial fauna
in Namibia, Africa
- Table 1: Significant results of GAMLSS models exploring the association between
ignorance scores and environmental and socio-geographical factors for the five
reference taxonomic groups in Namibia (p < 0.05). The complete results, including nonsignificant associations, are available in the Supplementary Material 2……………… 90

Sumário
Apresentação ................................................................................................................. 13
Revisão da Literatura ................................................................................................... 16
Referências .................................................................................................................. 30
Objetivos ........................................................................................................................ 37
Capítulo 1: : How taxonomic change influences forecasts of the Linnean Shortfall
(and what we can do about it)………………………………………………………...38
Abstract ....................................................................................................................... 39
Introduction ................................................................................................................. 40
References ................................................................................................................... 49
Capítulo 2: Do biological and ecological variables influence nomenclatural
uncertainty in birds?............................................…………………………………….55
Abstract ....................................................................................................................... 56
Introduction ................................................................................................................. 57
Methods....................................................................................................................... 59
Results ......................................................................................................................... 63
Discussion ................................................................................................................... 67
References ................................................................................................................... 71
Capítulo 3: Quantifying spatial ignorance in the effort to collect terrestrial fauna
in Namibia, Africa…………………………………………………………………….77
Abstract ....................................................................................................................... 78
Introduction ................................................................................................................. 79
Methods....................................................................................................................... 81
Results ......................................................................................................................... 86
Discussion ................................................................................................................... 91
References ................................................................................................................... 96
Conclusões gerais ........................................................................................................ 103

Apresentação
Espécie é a unidade taxonômica primordial para investigar os demais
componentes da biodiversidade. A partir do reconhecimento, descrição e nomeação das
espécies, que cientistas podem então conhecer sobre dados populacionais, taxas de
endemismo, status de conservação, quais as relações evolutivas, qual nicho e papel
ecológico daquela peça na complexidade da natureza. Faz parte da cultura humana
classificar a vida, e por isto ainda não existe um consenso universal sobre o que são
espécies e como nomeá-las. Muitos conceitos de espécies foram lançados na literatura,
sejam mais tradicionais baseando-se em caracteres morfológicos, ou mais sofisticados,
incluindo diversas fontes de informação. Como consequências, Espécies são
consideradas hipóteses, que ao serem reveladas para comunidade científica podem ou
não ser aceitas e adotadas.
Estas diversas fontes de informação, conceitos e métricas, podem - e acarretam em incertezas taxonômicas, que, por conseguinte, afetam as estimativas de riqueza de
espécies. Além disso, os taxonomistas, isto é, os cientistas que têm o papel de descrever
e nomear as espécies se baseiam em códigos internacionais de nomenclatura, que por
sua vez, permite liberdade na escolha dos nomes científicos. Isto também gera casos de
sinonímia, quando a mesma espécie foi nomeada com diferentes nomes científicos, ou
ainda processos de mudanças taxonômicas, quando as espécies são trocadas, agrupadas
ou divididas. A partir dos processos de revisões taxonômicas que acontece o
intercâmbio taxonômico (swap), ou seja, quando a espécie muda completamente de
táxon, o agrupamento (lumping) que é quando várias espécies (ou subespécies) são
consideradas parte da mesma espécie, ou o processo inverso, ocorrendo à divisão
(splitting) de uma espécie em várias diferentes.
Para a elaboração de planos e ações de conservação e para a avaliação de como a
biodiversidade está se adaptando às mudanças evolutivas e antrópicas na natureza, é
essencial que os dados e informações da biodiversidade sejam disponíveis e que estes
apresentem qualidade e confiabilidade suficientemente aplicável. Iniciativas nacionais e
internacionais têm aumentado o compartilhamento de dados em todo o mundo,
diminuindo assim fronteiras para o desenvolvimento de pesquisas. Atualmente, uma
iniciativa de grande relevância por disponibilizar mais de 2.6 bilhões de dados, tanto de
coleções museológicas quando de ciência cidadã, é o Global Biodiversity Information
Facility (GBIF). Entretanto, estes dados não são distribuídos de forma homogênea, e
13

existem muitas tendências e vieses, seja na distribuição espacial, taxonômica e
temporal. Diante dos desafios apontados, a presente tese foca nas lacunas Linneana e
Wallaceana, isto é, nos respectivos conhecimentos (ou escassezes) da taxonomia e
distribuição geográfica da biodiversidade.
A tese é dividida em duas partes, sendo a primeira parte uma revisão da
literatura sobre a lacuna Linneana, abrangendo os diversos conceitos de espécies e a
problemática por trás disto, e a lacuna Wallaceana, com foco nos vieses de recolha de
dados biológicos. A segunda parte trata-se de três capítulos em formato de manuscrito.
O primeiro capítulo é um manuscrito intitulado “How taxonomic change
influences forecasts of the Linnean Shortfall (and what we can do about it)” que foi
publicado na Journal of Biogeography. Elaboramos uma perspectiva sobre os
impedimentos que cientistas enfrentam para estimar a lacuna Linneana, ou seja,
desvendar o número de espécies existentes e descritas no mundo. Além disso,
discutimos que as previsões de riqueza de espécies são influenciadas tanto pelas taxas
de exploração e amostragem, como também podem ser superestimadas ou subestimadas
pelas mudanças taxonômicas, especialmente, as divisões e agrupamentos taxonômicos.
Finalmente, abordamos a importância de incluir estas informações nas métricas de
estimativas da biodiversidade para torná-las mais precisas e confiáveis, porém
reconhecendo que este tipo de conhecimento não é de amplo acesso nos bancos de
dados ou são mal documentados para maioria dos táxons.
No segundo capítulo, o manuscrito intitulado “Do biological and ecological
variables influence nomenclatural uncertainty in birds?”, abordamos a problemática da
lacuna Linneana quanto à nomenclatura das espécies. Precisamente, nós avaliamos as
várias listas de aves globais, que além de apresentarem números divergentes de
espécies, constam divergências de nomenclatura, isto é, espécies que apresentam nomes
científicos diferentes, seja nome genérico, nome específico e/ou nome completo.
Criamos e calculamos para 11.140 espécies de aves, uma métrica para avaliar a
incerteza de nomenclatura, utilizando a proporção de discordância (e ausência) de
nomes científicos entre as listas globais de aves. Posteriormente, nós avaliamos se e
quais características biológicas e ecológicas (massa corporal, tamanho da área de
distribuição, densidade do habitat, estilo de vida, status da IUCN e distinção evolutiva)
das aves estão associadas com a incerteza de nomenclatura.
O terceiro e último capítulo, possui o título “Quantifying spatial ignorance in
the effort to collect terrestrial fauna in Namibia, Africa” e foi publicado na Ecological
14

Indicators. Tratamos sobre a lacuna Wallaceana no sentido de avaliar as tendências e
lacunas temporais e espaciais no esforço amostral da biodiversidade. Como estudo de
caso, utilizamos a biodiversidade terrestre (aves, mamíferos, répteis, anfíbios e insetos)
da Namíbia por meio de dados do GBIF como estudo de caso, e quantificamos as
lacunas espaciais pela métrica de “Ignorance Score”. A biodiversidade da Namíbia é
extremamente diversificada e adaptada às condições áridas da região. Apesar da baixa
densidade populacional, a Namíbia apresenta um forte sistema de áreas protegidas, com
cerca de 40% do seu território protegido. Avaliamos quais variáveis sociogeográficas
(densidade rodoviária, densidade populacional humana, distância de instituições de
pesquisa, distância de área protegida e cobertura vegetal) influenciam as taxas de
ignorância espacial da biodiversidade. Finalizamos reforçando sobre usabilidade e
confiabilidade das recentes abordagens para avaliar qualidade de esforço amostral e
incertezas espacial, temporal e taxonômica.

15

Revisão da Literatura
Sistemas de Classificação dos Organismos
Caracterizar os organismos e descrever processos e padrões do mundo natural faz parte
da característica humana há milênios. Um notável exemplo foram os feitos dos filósofos
gregos Aristóteles (384 a.C. – 322 a.C.) e seu discípulo Teofrasto (372-287 a.C.).
Aristóteles criou uma classificação dos organismos vivos, separando-os pela presença
de sangue e similaridades morfológicas. Ele delimitou alguns grupos faunísticos, como
mamíferos, peixes, aves e insetos, embora tenha realizado alguns enquadramentos
errôneos, como não ter incluído as baleias aos mamíferos, devido à diferença no
formato corporal (KLEPKA; CORAZZA, 2018). Aristóteles acreditava que os
organismos deveriam ser classificados pela ordem crescente de complexidade. Já
Teofrasto elaborou uma classificação botânica, agrupando plantas de acordo com seus
habitats, periodicidade e porte, dividindo assim as herbáceas, os arbustos e as árvores, e
as estações anuais, bienais e perenes (GOMES-DA-COSTA, 2010). Estas ideias foram
perduradas por mais de 2.000 anos, até ser superada pela classificação sistematizada
proposta por Carl von Linné (Lineu) (1707 – 1778).
Na classificação de Lineu existiam cinco seções: classe, ordem, gênero, espécie
e variedade. Na seção gênero eram agrupadas as características morfológicas entre as
espécies, que eram obtidas em três diferentes caracteres: artificial, essencial e/ou natural
(KLEPKA; CORAZZA, 2018). O termo características essenciais era empregado para
descrever os atributos necessários que diferenciava uma espécie da outra. Seu sistema
não reflete necessariamente as relações entre os organismos e seu grau de parentesco.
Entretanto, sua trajetória foi marcada pela criação do sistema de binômios latinos para
os nomes das espécies de plantas e animais. Mais de 250 anos após Lineu iniciar seu
inventário sistemático das espécies do mundo, ainda há debates consideráveis sobre
quantas espécies realmente existem e quantas delas já foram documentadas. Isso
acontece devido à biodiversidade, complexidade e dinamismo da natureza, adicionado à
baixa capacidade humana de coletar e identificar as espécies (LADLE; HORTAL,
2013). Deste modo, alcançar uma compreensão abrangente sobre qualquer aspecto da
biodiversidade permanece em grande parte inalcançável.
O número estimado de espécies viventes em todo o mundo varia entre 3 e 100
milhões (MAY, 2010), baseado em opinião de especialistas, entretanto este número é
divergente ao depender da métrica utilizada (Tabela 1). Apenas uma pequena fração de
16

toda a biodiversidade foi formalmente descrita, cerca de 2 milhões de espécies (BÁNKI
et al., 2022), e muitas foram extintas antes mesmo de serem descobertas (COSTELLO;
MAY; STORK, 2013). Da mesma maneira, o conhecimento sobre a distribuição das
espécies é espacialmente heterogênea, com muitas áreas e ambientes mais conhecidos
que outros. Logo, porque algumas espécies e regiões receberam devida atenção e
empenho para uma avaliação mais completa do que outras? São características
intrínsecas ou extrínsecas às espécies que norteiam nosso conhecimento da vida? São
ambientes mais acessíveis que aumentam o conhecimento da sua biodiversidade? Tais
classificações refletem os objetivos de quem os fazem, ou seja, são agregados aspectos
culturais, temporais e espaciais, servindo como ilustrações para representar a realidade e
produzir conhecimento científico (ROSEN, 1996). Sendo assim, a mensuração da
biodiversidade (e seus aspectos taxonômicos, espaciais e temporais) constitui apenas
uma das múltiplas abordagens para enquadrar a diversidade da vida.
Lacunas do Conhecimento da Biodiversidade
Atualmente, a ausência ou carência de conhecimento sobre os aspectos da
biodiversidade são divididos em sete categorias de lacunas (BROWN; LOMOLINO,
1998; HORTAL et al., 2015). Estas lacunas representam importantes áreas da biologia,
e onde a pesquisa-ação é particularmente necessária. Resumidamente, são elas: a)
Lacuna Linneana (taxonomia): muitas espécies ainda não foram descritas ou
catalogadas; b) Lacuna Wallaceana (distribuição geográfica): falta de informações sobre
a distribuição de espécies em diferentes regiões geográficas; c) Lacuna Prestoniana
(ecologia de populações): falta conhecimento sobre os tamanhos populacionais,
dinâmicas e abundância das espécies; c) Lacuna Darwiniana (evolução): falta de
conhecimento sobre a evolução e árvore da vida das espécies; d) Lacuna Raunkiareana
(traços funcionais): desconhecimento sobre as funções e características ecológicas das
espécies; e) Lacuna Hutchinsoniana (tolerâncias abióticas): falta de conhecimento sobre
como as espécies se adaptam e toleram as mudanças ambientais; f) Lacuna Eltoniana
(interações ecológicas): falta de compreensão sobre como as espécies interagem e
sobrevivem.
Estas lacunas do conhecimento da biodiversidade significam que os cientistas
estão trabalhando muitas vezes com dados incompletos, incertos e não representativos e
sobre um número limitado de organismos e suas características (LADLE; HORTAL,
2013). Isto compromete, por exemplo, a capacidade de descrever a biodiversidade
17

existente, fazer previsões precisas sobre como os organismos estão sendo impactados
por mudanças ambientais e como podem se adaptar, e utilizar de forma ineficiente de
recursos para conservação. Todas as lacunas do conhecimento estão interconectadas em
graus variados, de acordo com a escala e a cobertura espacial, temporal e taxonômica
(HORTAL et al., 2015). Logo, reconhecer e quantificar as lacunas no conhecimento
sobre da biodiversidade global é imperativo (CARDOSO et al., 2011).
A Lacuna Linneana
A lacuna Lineana afeta criticamente todas as demais lacunas, pois a falta de
informação sobre a identidade das espécies impede necessariamente a descrição de
qualquer outra característica atribuída e relação envolvida (WHITTAKER et al., 2005).
Antes mesmo de existir qualquer estudo acerca da biologia, ecologia e/ou evolução da
biodiversidade, é fundamental reconhecer o grupo taxonômico em questão. A
taxonomia é a disciplina da biologia com o objetivo de identificar, descrever, classificar
e nomear os organismos vivos e extintos. Espécie é comumente a hierarquia avaliada
para atingir metas de conservação, como por exemplo, a proteção de espécies
endêmicas, emblemáticas e ameaçadas de extinção e os ambientes em que habitam
(AGAPOW et al., 2004; BARROWCLOUGH et al., 2016; LADLE; WHITTAKER,
2014).
Mesmo no século XXI, muitas espécies continuam sendo descobertas,
especialmente devido aos esforços de taxonomistas, avanços nas análises genéticas e
reavaliações de acervos museológicos (LADLE; WHITTAKER, 2014). Em um mundo
perfeito, cada espécie recém-descrita teria sido coletada, meticulosamente documentada,
identificada de forma inequívoca, nomeada e alocada permanentemente um ramo único
na árvore da vida, baseados em métodos idênticos e em um conceito universalmente
aplicado do que constitui uma espécie (STROPP et al., 2022). Entretanto, no mundo
real, os taxonomistas têm usado uma variedade de conceitos de espécies e abordagens
(relevantes, porém limitantes) para identificar novas espécies e reavaliar espécies já
documentadas (Tabela 1) (KITCHENER et al., 2022; ZACHOS, 2016).

18

Tabela 1: Principais abordagens para estimar a lacuna Linneana (modificado de MORA et al.,
2011); as limitações foram descritas a partir da análise das abordagens pela autora desta tese e
seu orientador.
Abordagem
Descrição e referência
Limitação
Multiplicação das proporções relativas de
espécies que são conhecidas de ocorrer
A estimativa não é confiável
Índices de
em um habitat para estimar os números
para grupos taxonômicos
diversidade
prováveis de ocorrer em outros habitats
hiperdiversos e pouco
(ERWIN, 1982; GARCÍA-ROBLEDO et
estudados.
al., 2020; HAMILTON et al., 2010)
Extrapolando a partir de taxas históricas
As variações temporais no
Curvas de
de descoberta de espécies (BEBBER et
processo de descoberta tornam
descoberta
al., 2007)
as curvas pouco informativas.
Há grande incerteza sobre as
Suposições fundamentadas ou pesquisas
estimativas, sem protocolos,
Opinião de
de opinião de especialistas, como o
variando para mais ou menos.
especialistas
método Delphi (FISHER et al., 2015)
Interessante apenas para
táxons novos.
Extrapolação de proporções de espécies
O desempenho do estimador
Extrapolação
não descritas de uma área intensamente
deteriora-se à medida que o
de área bem
estudada para a extensão total dessa área
tamanho da amostra prevista
amostrada
(SHEN; CHAO; LIN, 2003)
aumenta.
As estimativas são confiáveis
Extrapolação a partir de proporções de
para mudanças taxonômicas
Extrapolação
números de espécies em categoria
em táxons superiores. A
de táxons bem taxonômica alta para prever os números
interpretação da extrapolação
conhecidos
em áreas onde apenas alguns táxons foram
no táxon inferior deve ser feita
bem descritos (MORA et al., 2011)
com cautela.
As estimativas utilizam apenas
Relação
Extrapolação das relações entre tamanhos
espécies conhecidas.
tamanho
corporais e números de espécies
Incomparável entre grupos
corporal
(ZAPATA; ROSS ROBERTSON, 2007)
taxonômicos.
Em regiões pouco exploradas,
Estimando a riqueza de espécies de uma
muitos táxons são
Relação
grande área a partir de pesquisas locais
consequentemente mal
espécie-área
dispersas dentro dela (KUNIN et al.,
amostrados, tornando inviável
2018)
o acúmulo de informações.
Extrapolação da razão entre espécies
Muitas fontes de erros estão
aceitas e sinônimos observados para
associadas a dados
Taxas de
famílias de plantas individuais
taxonômicos, tornando-os
Sinonímia
(GOVAERTS, 2001; SCOTLAND;
pouco confiáveis. Não utiliza
WORTLEY, 2003)
dados históricos.
Existem lacunas de dados
Estimativas de
Usando atributos biológicos, ambientais e multivariados para muitos
probabilidades
sociológicos em nível de espécie para
táxons. O modelo não é capaz
de descoberta
avaliar a probabilidade de descoberta de
de distinguir definições de
em nível de
espécies (MOURA; JETZ, 2021)
espécies válidas e mudanças
espécie
nas práticas taxonômicas.
Usando machine learning para criar um
Não existe um banco de dados
Extrapolações modelo preditivo que identifica espécies
de sequências genéticas para
de níveis de
nomeadas que provavelmente contêm
muitos táxons. Ainda é
amostragem
diversidade oculta com base nos níveis de
necessário revisar muitas
molecular
cobertura de dados genéticos (PARSONS
sequências problemáticas.
et al., 2022)
19

São dois grandes desafios que a taxonomia moderna enfrenta: o primeiro é
encontrar um consenso sobre a definição de espécies, e o segundo é delimitar e
descrever o número de espécies existente em todo o mundo (PADIAL et al., 2010). A
verdade é que espécies são hipóteses, e frequentemente seu conceito é baseado em
padrões de similaridade, ancestralidade, evolução e filogenia (PANTE et al., 2015). Em
1997, Mayden realizou um levantamento sobre os conceitos teóricos de espécies
(MAYDEN, 1997), e encontrou 24 conceitos, cujas incompatibilidades de definição
poderiam gerar conclusões divergentes quanto ao número e limites do que é espécie.
Os conceitos de espécies mais tradicionais na literatura são: a) Conceito
biológico: espécie é um grupo de organismos que são capazes de se reproduzir entre si,
produzindo descendentes férteis, e que estão reprodutivamente isolados de outros
grupos. Isso significa que os membros de uma espécie podem se acasalar e gerar
descendentes viáveis, enquanto são isolados de outros grupos reprodutivamente
(MAYR; PROVINE, 1980); b) Conceito ecológico: espécie é definida como uma
linhagem que ocupa uma zona adaptativa (ou seja, determinadas pelos recursos
explorados e hábitats ocupados) divergente de outras linhagens, na sua área de
distribuição, e que evolui separadamente de todas as linhagens (VAN VALEN, 1976);
c) Conceito evolutivo: uma espécie é vista como uma linhagem única de organismos
que compartilham características derivadas de seu ancestral comum e que mantém suas
características separada de outras linhagens, como adaptações específicas ao ambiente,
modificações genéticas ou outras características que evoluíram ao longo do tempo
(WILEY, 1978); d) Confeito Filogenético: espécie é considerada como o menor grupo
de um conjunto de organismos que compartilham um padrão parental de ancestralidade
e descendência, isto é, um ancestral comum exclusivo (monofilético) (CRACRAFT,
1983).
Na prática os diferentes conceitos de espécies têm consequências importantes na
biologia e conservação (DE QUEIROZ, 2007). Por exemplo, podem alterar as
estimativas de riqueza de espécies. O conceito filogenético de espécies, baseado em
ancestralidade, tem sido apontado por reconhecer um maior número de espécies quando
comparado ao conceito biológico, baseado em morfologia. Um estudo realizado em
2005 avaliou as implicações da aplicação de diferentes conceitos de espécies nas
estimativas de riqueza de aves da África subsaariana, e revelou que houve um aumento
de 33% no número de espécies quando utilizado o conceito de espécies filogenéticas
(n= 2.098) em relação ao conceito de espécies biológicas (n= 1.572) (DILLON;
20

FJELDSÅ, 2005). O mesmo padrão foi observado para aves do México (NAVARROSIGÜENZA; PETERSON, 2004; PETERSON; NAVARRO-SIGÜENZA, 1999). Além
disso, os diferentes conceitos de espécies podem afetar estudos de história natural,
padrões de fluxo gênico e avaliações de áreas geográficas, como a delimitação de áreas
de endemismos, áreas de alta biodiversidade e área prioritárias para conservação
(BATES; DEMOS, 2001; MEIJAARD; NIJMAN, 2003; PETERSON; NAVARROSIGÜENZA, 1999). Portanto, identificações erradas de espécies nos levam a respostas
erradas ou inconsistentes para questões da biologia e conservação.
A taxonomia tradicional, baseada em morfoespécies, não leva em consideração a
biodiversidade críptica, isto é, espécies (ou complexo de espécies) que são
extremamente semelhantes na morfologia, porém apresentam distinções genéticas e/ou
ecológicas (SHIN; ALLMON, 2023). Apesar de não existir um consenso taxonômico
universal, e taxonomistas de diferentes áreas usam diversas práticas para descrever
espécies, já está sendo notado que o estudo da especiação é crucial para a taxonomia. As
espécies serão mais precisamente delimitadas quando entendermos as causas de sua
origem e as influências em suas trajetórias evolutivas, portanto, é necessário adotar uma
perspectiva múltipla e complementar neste processo (PADIAL et al, 2010).
Nas últimas décadas, houve um aumento expressivo nas taxas globais de
descrição para muitos grupos de organismos, especialmente devido aos avanços de
novas técnicas computacionais e ao acesso de dados museológicos, geográficos e
genéticos, (TROUDET et al., 2017). A taxonomia integrativa surge assim como
arcabouço para resolver os conflitos entre delimitações de espécies e revelar a
biodiversidade oculta de forma mais rápida e confiável do que os métodos tradicionais
(PADIAL et al., 2010). Nesta abordagem são utilizadas múltiplas fontes de dados
complementares

(morfológicos,

moleculares,

ecológicos,

comportamentais

e

geográficos) para descrever espécies, buscando uma compreensão mais holística das
relações filogenéticas e taxonômicas (DAYRAT, 2005). A taxonomia integrativa é
particularmente útil para resolver problemas taxonômicos complexos, como a
delimitação de espécies crípticas (espécies que são morfologicamente semelhantes, mas
geneticamente distintas), a identificação de populações em processo de especiação e a
revisão de grupos taxonômicos mal definidos.
A nomenclatura é o sistema pelo qual os taxonomistas se baseiam para nomear e
classificar as espécies/táxons. Para isto, os taxonomistas de diversas áreas utilizam as
regras estabelecidas pelos seus respectivos Códigos internacionais de nomenclatura, que
21

por sua vez, permitem liberdade quanto à escolha dos critérios e atributos, aumentando
incertezas e alterações taxonômicas (RHEINDT et al., 2023). A incerteza taxonômica
refere-se à falta de clareza ou confiança na classificação de organismos em categorias
taxonômicas específicas, como gênero, espécie e/ou família. À medida que novas
evidências surgem, especialmente através de estudos genéticos e filogenéticos, a
classificação taxonômica de certos grupos de organismos pode ser revisada. Isso pode
levar a mudanças na taxonomia que geram incerteza sobre a classificação anterior
(LESSA et al., 2024).
A mobilização (ou a falta) de dados também é um dos fatores que afeta a
incerteza taxonômica. Para alguns grupos taxonômicos, pode haver uma escassez de
dados disponíveis, sejam morfológicos, moleculares ou ecológicos, o que torna difícil
uma classificação precisa. Além disso, uma proporção das espécies que foram
formalmente descritas ainda não foi incluída em monografias taxonômicas ou listas de
verificação com curadoria (HORTAL et al., 2019). As listas de verificação de espécies
aceitas, cujo principal exemplo é o Catálogo da Vida (HOBERN et al., 2021), procuram
fornecer um consenso sobre todas as espécies válidas (e outras categorias de
classificação taxonômica superior e inferior). A descrição de uma espécie em periódico
especializado pode gerar um processo de discussão e revisão por parte da comunidade
taxonômica que inevitavelmente gera um lapso de tempo antes de sua aceitação geral e
da inclusão nessas listas (Figura 1).
Além disso, pode gerar potenciais divergências se o consenso diferir entre
diferentes comunidades científicas que utilizam conceitos de espécie diferentes e, assim,
mantêm listas de verificação diferentes para o mesmo táxon (NEKOLA; HORSÁK,
2022). Por exemplo, mesmo em grupos taxonômicos bem estudados, como as aves, as
lacunas e incertezas taxonômicas são proeminentes. Como reflexo, existem múltiplas
listas de aves globais, com base em múltiplos critérios taxonômicos, e múltiplos
propósitos científicos e conservacionistas (McCLURE, 2020; NEATE-CLEGG, 2021).
Incertezas taxonômicas podem levar a estimativas equivocadas da riqueza de espécies.
Um estudo recente descobriu que 68% de quatro listas de aves apresentam algum grau
de discordância, seja por omissões taxonômicas, uso de nome científico diferente (nome
completo, epíteto e/ou gênero) ou tratamento em nível taxonômico (espécie e
subespécie) (McCLURE, 2020). Portanto, lidar com este desafio é crucial e requer
compromissos com a conceituação das espécies.

22

Figura 1: Proporção de espécies atualmente aceitas em relação ao total de descrições
taxonômicas feitas a cada ano; As tonalidades das cores indicam o número de descrições
taxonômicas feitas a cada ano em escala logarítmica, sendo que as tonalidades mais claras
indicam maior número de descrições. Os painéis a) e b) representam dados de duas famílias de
plantas, Fagaceae e Solanaceae, enquanto os painéis c) e d) o fazem para duas famílias de
peixes, Cychlidae e Characidae. Fontes de dados: A lista de nomes de táxons, status taxonômico
e ano de publicação foi recuperada de Govaerts (2022) para as famílias de plantas, e de Froese
et al. (2022) para as famílias de peixes. Créditos da imagem: silhuetas foram baixadas em
https://thenounproject.com; Fagaceae (carvalho de Eucalyp); Solanaceae (tomateiro de Michael
Zick Doherty); Cychlidea (Cichlid por Ametyst Studio); Characidae (piranha de Agne Alesiute).
Figura elaborada pela Dra. Juliana Stropp, compartilhada com permissão. Os dados e o código
R estão disponíveis em https://github.com/justropp.

Existem pelo menos duas dimensões relacionadas à incerteza taxonômica: a) o
erro taxonômico: onde a identificação da espécie foi feita utilizando múltiplas
taxonomias e/ou múltiplos conceitos de espécies, espécies que passaram por revisões
taxonômicas, espécies em táxons mal resolvidos (complexo de espécies); b) o erro de
identificação: o identificador/taxonomista fez uma identificação incorreta quanto ao
nome científico, ou seja, pode depender da experiência do identificador (amador,
23

especialista ou taxonomista; TESSAROLO et al., 2021), do tipo de dado (fotografias,
observações humanas ou espécimes de museu), pode ocorrer em espécies simpátricas ou
ainda táxons mais recentes.
As incertezas taxonômicas mais comuns são relacionadas à sinonímia, isto é,
quando o mesmo táxon (por exemplo, espécie) possui nomes científicos diferentes
(MORA et al., 2011). Isso ocorre quando: i) taxonomistas de regiões diferentes estão
trabalhando de forma independente e publicam novas espécies sem o conhecimento
prévio; e ii) espécies estão “ocultas” dentro de outras espécies e são reveladas após de
revisões taxonômicas (DAYRAT, 2005). Neste último contexto, as revisões
taxonômicas podem gerar divisões (splitting), quando uma espécie é separada em mais
de uma espécie ou subespécies, ou o oposto, gerando agrupamentos (lumping), quando
várias espécies que eram reconhecidas como distintas são na verdade a mesma espécie.
Estes processos de revisões e alterações taxonômicas podem levar a inflação ou
deflação taxonômica, isto é, o aumento ou diminuição do número de espécies. Tem
havido intensos debates sobre se a inflação/deflação taxonômica é real, se é
simplesmente um reflexo do progresso taxonômico, e se influenciará significativamente
as percepções dos padrões de biodiversidade (GARNETT; CHRISTIDIS, 2017;
HARRIS; FROUFE, 2005; SANGSTER, 2009; WHEELER, 2020). Por exemplo, as
estimativas do número de espécies de palmeiras neotropicais (gênero Attalea) eram
muito maiores na década de 1970, antes que muitas espécies anteriormente válidas
fossem agrupadas (HENDERSON, 2020) (Figura 2).
Está cada vez mais evidente que o preenchimento desses déficits é extremamente
desafiador devido à existência de vários impedimentos, como a escassez de
taxonomistas para muitos táxons (ENGEL et al., 2021), e uma série de outros
obstáculos técnicos à coleta, compilação e análise de múltiplas formas de dados sobre a
biodiversidade. Em particular, os dados de ocorrência de espécies derivados de
espécimes armazenados em coleções museológicas sofrem de qualidade incerta e vieses
espaciais e temporais (STROPP et al., 2016). Além disso, os dados sobre alterações
taxonômicas não estão atualmente facilmente disponíveis para a grande maioria das
espécies, e quando disponíveis é preciso muito tempo para compilação e processamento
(KONSTANTINOV; NAMYATOVA, 2019), o que torna a análise das tendências
taxonômicas históricas altamente desafiadoras e acrescentando incerteza a muitos
métodos de previsão para avaliar a Lacuna Linneana (STROPP et al., 2022). Mesmo
que todos os taxonomistas concordassem em usar exclusivamente um único conceito de
24

espécie, ainda haveria divergências sobre a delimitação das espécies, uma vez que existe
um elemento de julgamento humano ao decidir se uma ou mais populações merecem ser
designadas como espécie (ZACHOS, 2018).

Figura 2: Mudança taxonômica em palmeiras amazônicas (Arecaceae; Palmae). O painel (a)
mostra a reclassificação taxonômica de 708 sinônimos em 148 nomes aceitos; cada cor
representa um nome de espécie. O painel (b) descreve detalhadamente a ligação entre sinônimos
heterotípicos e nome aceito em uma linha do tempo de agrupamento taxonômico (lumping) para
um desses 148 nomes aceitos, Attalea butyracea; linhas coloridas horizontais marcam o ano de
descrição de 18 sinônimos heterotípicos e o ano de sinonimização, sendo que cada cor
representa um sinônimo atual; as linhas verticais indicam a contagem de nomes de espécies
aceites num determinado ano; linhas curvas representam o agrupamento de 18 sinônimos
heterotípicos em A. butyracea. Fontes de dados: para o painel (a) a lista de nomes aceitos de
palmeiras amazônicas foi extraída de (Cardoso et al., 2017; ter Steege et al., 2019), enquanto a
lista de sinônimos foi obtida de (Govaerts et al., 2022); e a lista de sinônimos heterotípicos
mostrada no painel (b) foi obtida de Henderson (2020). Figura publicada em short
communication no Journal of Biogeography (https://doi.org/10.1111/jbi.14463), compartilhada
com permissão dos autores.

A ciência já te avançado para preencher as lacunas Linneana e reduzir as
incertezas taxonômicas. Como por exemplo, debatendo sobre os conceitos e definições
de espécies, utilizando diversas fontes de dados para fortalecer a hipótese de espécie,
como também, compartilhando em banco de dados digitais informações taxonômicas,
como descrições e listas de espécies e revisões taxonômicas. Entretanto, ainda há muito
trabalho e esforços a serem enfrentados. As estimativas atuais demonstram que estamos
400 anos atrasados em pesquisas taxonômicas, para obtenção de um inventário
completo (PADIAL et al., 2010). Considerando que as intensas atividades e pressões
25

humanas sobre a natureza estão levando à extinção de espécies, é urgente reconhecer a
biodiversidade existente, para revelar seus valores. Portanto, para melhor a qualidade e
previsões taxonômicas é necessário um esforço conjunto de taxonomistas e
macroecologistas para: i) documentar e descrever a história taxonômica das espécies,
incluindo informações sobre divisões, agrupamentos e revisões; ii) incorporar estas
informações em modelos de estimativas de riqueza de espécies; e iii) prever como estas
mudanças taxonômicas poderão remodelar os padrões de riqueza de espécies (STROPP
et al., 2022).
A Lacuna Wallaceana
O conhecimento sobre a taxonomia e distribuição geográfica das espécies é estritamente
conectado e considerado de fundamental importância para estudos da biodiversidade
(BINI et al., 2006; HORTAL et al., 2015). A falta de conhecimento sobre a distribuição
geográfica das espécies é conhecida como lacuna Wallaceana (HORTAL et al., 2015;
LOMOLINO, 2004). Esta lacuna é impulsionada pela variação temporal e espacial no
esforço de amostragem. A variação espacial no esforço amostral é heterogênea
resultando em áreas – sejam países, regiões ou ecossistemas – mal representadas em
coleções científicas e em bancos de dados de biodiversidade (LOBO, 2008). Isto ocorre
especialmente em áreas remotas, como densas florestas tropicais ou zonas abissais,
áreas montanhosas ou regiões de clima extremamente árido, que são consideradas
historicamente

negligenciadas

por

taxonomistas

e

biogeógrafos

(LADLE;

WHITTAKER, 2014; LESSA et al., 2019).
Além disso, existe viés nos dados de distribuição de espécies em termos de
conveniência/comportamento dos cientistas (e coletadores) e das tendências históricas
de colonização e inventariação (BINI et al., 2006; MEYER et al., 2015; SASTRE;
LOBO, 2009). Estes vieses espaciais podem ocorrer dentro de unidades políticas, com
tendência de uma maior concentração de registros de ocorrências de espécimes em áreas
mais acessíveis, localizados perto das vias de acesso, como estradas e rios navegáveis,
em áreas conhecidas por serem ricas em espécies, como as áreas protegidas, áreas
próximas às instituições de pesquisas e em cidades com melhor infraestrutura
(SASTRE; LOBO, 2009). Por exemplo, dentro do continente africano há uma
disparidade na distribuição espacial de registro de ocorrência de plantas com flores,
sendo a África do Sul apresentando trinta vezes mais dados quando comparado com
países vizinhos, como a Namíbia (STROPP et al., 2016). Em resumo, áreas remotas,
26

inacessíveis ou politicamente sensíveis podem não ser adequadamente conhecidas
quanto sua biodiversidade. Esta desigualdade no esforço amostral da biodiversidade
pode fazer com que a interpretação de mapas de riqueza de espécies seja similar aos
mapas de esforço amostral, um padrão que é visualmente marcante na África
Subsaariana (LESSA et al., 2024; STROPP et al., 2016).
A variação espacial no esforço amostral está, por sua vez, relacionada com a
variação na disponibilidade de recursos humanos e financeiros, e da capacidade
científica da área de estudo (RUETE, 2015). A lacuna Wallaceana é mais proeminente
no sul global, devido à sua dependência das tendências históricas em capacidade
científica e recursos humanos (BECK et al., 2014; HORTAL et al., 2015; JETZ;
MCPHERSON; GURALNICK, 2012). Coleções museológicas localizadas em cidades
desenvolvidas tendem a receber mais material biológico de diversos lugares (Penn,
Cafferty, & Carine, 2018), ou ainda pesquisadores encontram melhores infraestruturas
(universidades e laboratórios) e recursos financeiros em centros urbanos (CORREIA et
al., 2019; KADMON; FARBER; DANIN, 2004; LESSA et al., 2019; LOBO, 2008;
OLIVEIRA et al., 2016). A coleta de dados sobre a distribuição de espécies muitas
vezes requer recursos significativos, incluindo financiamento, tempo de atividade de
campo e pesquisadores qualificados. Em algumas áreas, especialmente em países em
desenvolvimento, pode haver falta de capacidade técnica para realizar levantamentos
biológicos abrangentes, falta de acesso a tecnologias e métodos eficazes de coleta de
dados, o que contribui para a falta de informações detalhadas sobre a distribuição de
espécies (LESSA et al., 2024).
O cenário passa a ser mais crítico devido à rápida transformação dos ambientes
naturais em detrimento do uso humano (como desmatamentos para agricultura,
mineração, produção de energia, transporte e construção civil), às alterações climáticas
globais, desastres ambientais, e até mesmo perdas físicas (tal como incêndios em
coleções museológicas), que têm afetado a qualidade dos dados de distribuições de
espécies ao longo do tempo, tornando-os imprecisos e desatualizados (LADLE;
HORTAL, 2013; LADLE; WHITTAKER, 2011; NABOUT et al., 2016; TESSAROLO
et al., 2017; VEACH; MOILANEN; MININ, 2017). Em respostas às mudanças
ambientais as espécies podem alterar suas dinâmicas espaciais em curta escala temporal,
diminuindo a qualidade e a acurácia dos dados de distribuição mais antigos (HUGHES
et al., 2012; STROPP et al., 2020). Isto pode resultar na perda de conhecimento das
populações das espécies antes mesmo de serem documentadas. Como demonstrado no
27

estudo publicado em 2017 que avaliou como as taxas de desmatamento na Amazônia
brasileira afetariam os dados de distribuição de espécies de árvores. Os autores
descobriram que até o ano de 2017, 12% da Amazônia brasileira foi desmatada, sem que
houvesse um único exemplar de árvore registrada em coleções museológicas, e 37% das
áreas apontadas como bem amostradas teriam sido desmatadas (STROPP et al., 2020).
Portanto, áreas desmatadas em ambientes inexplorados tem o poder de dizimar a
biodiversidade, bem como prejudicar a oportunidade de investigação e recolha de dados
em campo.
A desigualdade dos dados de distribuição de espécies apresentam consequências
inestimáveis para a conservação das espécies, uma vez que estes dados são usados para
identificação e priorização de áreas para conservação (KUJALA; MOILANEN;
GORDON, 2018) e para previsão de impactos de futuras mudanças ambientais
(SINGER et al., 2016; THUILLER et al., 2008). A lacuna Wallaceana tem um impacto
importante nas estimativas dos status de ameaça e risco de extinção, pois para
elaboração de planos e ações de conservação são necessárias informações acuradas
sobre o tamanho da área de distribuição das espécies-alvo (HORTAL et al., 2015). A
União Internacional para Conservação da Natureza (IUCN) utiliza esta informação para
definir e classificar o status de ameaça das espécies, sendo consideradas prioritárias as
espécies com menor distribuição geográfica (RIDDLE et al., 2011; RODRIGUES et al.,
2006). A falta de conhecimento sobre a distribuição das espécies também afetar os
parâmetros utilizados para quantificar a biodiversidade, como estimativas de riqueza de
espécies, delimitar o nicho das espécies, identificar o grau de endemismo, entre outras
(COSTELLO; WILSON, 2011; MORA et al., 2011; RIDDLE et al., 2011; ROCCHINI
et al., 2011; WHITTAKER et al., 2005).
Todos estes vieses espaciais e problemas relacionados corroboram para um
aumento da incerteza e redução da qualidade dos dados de distribuições de espécies.
Uma forma de mitigar o efeito da lacuna Wallaceana, é tornar os dados de distribuição
de espécies intactos e acessíveis em todo o mundo, para isso é fundamental digitalizar e
manter estes dados em plataformas digitais (HEDRICK et al., 2020). Em 2001, foi
implementada uma rede internacional de compartilhamento de dados de espécimes de
coleções de história natural e de ciência cidadã, o Global Biodiversity Information
Facility (GBIF) (BECK et al., 2013; CHANDLER et al., 2017; GAIJI et al., 2013;
NELSON; ELLIS, 2019). Atualmente o GBIF é a maior base de dados da
biodiversidade em todo o mundo, abrigando mais de 2.6 bilhões de registros de espécies
28

de vários grupos taxonômicos, desde indivíduos unicelulares a vertebrados e plantas
(GBIF, 2023). Desta forma, investigadores de qualquer lugar do mundo podem acessar
informações disponíveis na internet, ultrapassando barreiras geográficas para elaboração
de pesquisas acerca da biodiversidade.
Entretanto, mesmo grandes bancos de dados da biodiversidade, como o GBIF,
sofrem de incompletude, isto é faltam dados completos ou confiáveis, seja da
taxonomia, do ano de coleta ou da distribuição espacial (DE ARAUJO; QUARESMA;
RAMOS, 2022; FREEMAN; PETERSON, 2019; ROCHA‐ORTEGA; RODRIGUEZ;
CÓRDOBA‐AGUILAR, 2021). Isso acontece, muitas vezes, porque os dados de
ocorrência de espécies possuem diversas fontes de informação. O primeiro passo para
obter um conhecimento de qualidade da biodiversidade é reconhecer os limites do
conhecimento atual, identificando o quanto não sabemos e as lacunas relacionadas. Uma
vez que tenhamos descrito suficientemente o conhecimento da biodiversidade e seus
vieses e limitações, a próxima tarefa é melhorar o inventário da pesquisa global sobre a
biodiversidade de forma a maximizar a cobertura espacial e dispor de maneira mais
eficaz os recursos limitados para pesquisa e conservação. A criação de “mapas da
ignorância” é um das estratégias utilizadas para fornecer uma medida completa da
confiabilidade dos dados de ocorrência de espécies (CORREIA et al., 2019; RUETE,
2015; STROPP et al., 2016; TESSAROLO et al., 2021).
Existem várias abordagens que avaliam as lacunas e qualidade do conhecimento
da biodiversidade. Uma das formas de fazer isso é através da criação de “mapas de
ignorância” que distinguem áreas com amostragem intensiva daquelas com amostragem
insuficiente (ROCCHINI et al., 2011). Uma medida para criar tais mapas é através do
“Inventory Completeness”, que calcula a integridade dos inventários de espécies a partir
de repositórios digitais em diferentes unidades espaciais (SOUSA-BAENA; GARCIA;
PETERSON, 2014; STROPP et al., 2016), estimando curvas de acumulação de espécies
em cada unidade amostral com base no número de registros. No entanto, esta
abordagem não funciona bem para inventários de espécies não padronizados, e é muito
sensível a números baixos de registos, particularmente em regiões com elevada
biodiversidade (CORREIA et al., 2019; RUETE, 2015).
Outra abordagem ainda mais robusta é a criação de ‘Maps of biogeographical
ignorance’ (MoBIs), que utiliza diversas fontes de informação, tal como o “Inventory
Completeness”, a qualidade taxonômica, e o decaimento temporal e espacial dos dados
da biodiversidade (HORTAL et al., 2022; TESSAROLO et al., 2021). Por ser mais
29

elaborada esta métrica também é sensível ao baixo número de registros de ocorrência e
a baixa abundância de espécies em coleções de história natural (Meyer et al., 2016,
Steege et al., 2011, Stropp et al., 2016). Uma abordagem alternativa mais simples que as
anteriores é o “Ignorance Score”. As pontuações de ignorância da biodiversidade tem a
vantagem de ser simples de calcular, não utiliza estimativas de riqueza de espécies
(como as curvas de acumulação), e baseia-se apenas em dados brutos de ocorrências de
espécies (CORREIA et al., 2019; LESSA et al., 2024; MAIR; RUETE, 2016; RUETE,
2015), sendo ideal para ser aplicados em áreas com pouco conhecimento disponível.
Como resultado são fornecidas informações sobre a cobertura e confiabilidade da
amostragem, e relata explicitamente a distribuição espacial do viés e a falta de esforço
de amostragem em uma região de estudo.
Independente da abordagem utilizada é necessário reconhecer que a qualidade
dos dados é crucial para resultados fidedignos da distribuição espaciais, não só da
biodiversidade, mas dos esforços de cientistas e coletadores em explorar a natureza. O
preenchimento da lacuna Wallaceana contribui significativamente para o entendimento
da ecologia e da biogeografia, mostrando como fatores históricos e ecológicos moldam
a distribuição e a diversidade das espécies ao longo do tempo.
Referências
AGAPOW, P. et al. The Impact of Species Concept on Biodiversity Studies. The Quarterly
Review of Biology, v. 79, n. 2, p. 161–179, jun. 2004.
BÁNKI, O. et al. Catalogue of life checklist (Version 2022-01-14).
BARROWCLOUGH, G. F. et al. How many kinds of birds are there and why does it matter?
PLoS One, v. 11, n. 11, p. e0166307, 2016.
BATES, J. M.; DEMOS, T. C. Do we need to devalue Amazonia and other large tropical
forests? Diversity and Distributions, v. 7, n. 5, p. 249–255, 20 set. 2001.
BEBBER, D. P. et al. Predicting unknown species numbers using discovery curves. Proceedings
of the Royal Society B: Biological Sciences, v. 274, n. 1618, p. 1651–1658, 2007.
BECK, J. et al. Online solutions and the “Wallacean shortfall”: What does GBIF contribute to
our knowledge of species’ ranges? Diversity and Distributions, v. 19, n. 8, p. 1043–1050, 2013.
BECK, J. et al. Spatial bias in the GBIF database and its effect on modeling species’ geographic
distributions. Ecological Informatics, v. 19, p. 10–15, 2014.
BINI, L. M. et al. Challenging Wallacean and Linnean shortfalls: Knowledge gradients and
conservation planning in a biodiversity hotspot. Diversity and Distributions, v. 12, n. 5, p. 475–
482, 2006.

30

BROWN, J. H.; LOMOLINO, M. V. Biogeography. Sunderland, MA: [s.n.].
CARDOSO, P. et al. The seven impediments in invertebrate conservation and how to overcome
them. Biological Conservation, v. 144, n. 11, p. 2647–2655, nov. 2011.
CHANDLER, M. et al. Contribution of citizen science towards international biodiversity
monitoring. Biological Conservation, v. 213, p. 280–294, set. 2017.
CORREIA, R. A. et al. Using ignorance scores to explore biodiversity recording effort for
multiple taxa in the Caatinga. Ecological Indicators, v. 106, n. June 2019, p. 105539, 2019.
COSTELLO, M. J.; MAY, R. M.; STORK, N. E. Can We Name Earth’s Species Before They
Go Extinct? Science, v. 339, n. 6118, p. 413–416, 25 jan. 2013.
COSTELLO, M. J.; WILSON, S. P. Predicting the number of known and unknown species in
European seas using. Global Ecology and Biogeography, n. 20, p. 319–330, 2011.
CRACRAFT, J. Species concepts and speciation analysis. Current ornithology. Vol. 1, p. 159–
187, 1983.
DAYRAT, B. Towards integrative taxonomy. Biological Journal of the Linnean Society, v. 85,
n. 3, p. 407–415, 24 jun. 2005.
DE ARAUJO, M. L.; QUARESMA, A. C.; RAMOS, F. N. GBIF information is not enough:
national database improves the inventory completeness of Amazonian epiphytes. Biodiversity
and Conservation, v. 31, n. 11, p. 2797–2815, 2022.
DE QUEIROZ, K. Species Concepts and Species Delimitation. Systematic Biology, v. 56, n. 6,
p. 879–886, 1 dez. 2007.
DILLON, S.; FJELDSÅ, J. The implications of different species concepts for describing
biodiversity patterns and assessing conservation needs for African birds. Ecography, v. 28, n. 5,
p. 682–692, 20 out. 2005.
ENGEL, M. S. et al. The taxonomic impediment: a shortage of taxonomists, not the lack of
technical approaches. Zoological Journal of the Linnean Society, v. 193, n. 2, p. 381–387, 24
set. 2021.
ERWIN, T. L. Tropical forests: Their richness in Coleoptera and other arthropod species. The
Coleopterists Bulletin, v. 36, n. 1, p. 74–75, 1982.
FISHER, R. et al. Species richness on coral reefs and the pursuit of convergent global estimates.
Current Biology, v. 25, n. 4, p. 500–505, 2015.
FREEMAN, B.; PETERSON, A. T. Completeness of digital accessible knowledge of the birds
of western Africa: Priorities for survey. Condor, v. 121, n. 3, 2019.
GAIJI, S. et al. Content assessment of the primary biodiversity data published through GBIF
network: Status, challenges and potentials. Biodiversity Informatics, v. 8, n. 2, 9 jul. 2013.
GARCÍA-ROBLEDO, C. et al. The Erwin equation of biodiversity: From little steps to quantum
leaps in the discovery of tropical insect diversity. Biotropica, v. 52, n. 4, p. 590–597, 2020.
GARNETT, S. T.; CHRISTIDIS, L. Taxonomy anarchy hampers conservation. Nature News, v.
546, n. 7656, p. 25, 2017.
31

GBIF. GBIF Home Page. Disponível em: https://www.gbif.org/grscicoll/institution/search>.
Acesso em: 7 jan. 2023.
GOMES-DA-COSTA, S. M. Sistemas de Classificação em Reinos da Diversidade Biológica.
In: Organização dos Seres Vivos. Maringá: Eduem, 2010. p. 55–78.
GOVAERTS, R. How Many Species of Seed Plants Are There? Taxon, v. 50, n. 4, p. 1085–
1090, 2001.
HAMILTON, A. J. et al. Quantifying Uncertainty in Estimation of Tropical Arthropod Species
Richness. The American Naturalist, v. 176, n. 1, p. 90–95, jul. 2010.
HARRIS, D. J.; FROUFE, E. Taxonomic inflation: species concept or historical geopolitical
bias? Trends in Ecology & Evolution, v. 20, n. 1, p. 6–7, 2005.
HEDRICK, B. P. et al. Digitization and the Future of Natural History Collections. BioScience,
v. 70, n. 3, p. 243–251, 1 mar. 2020.
HENDERSON, A. J. A revision of Attalea (Arecaceae, Arecoideae, Cocoseae, Attaleinae).
Phytotaxa, v. 444, p. 1–76, maio 2020.
HOBERN, D. et al. Towards a global list of accepted species VI: The Catalogue of Life
checklist. Organisms Diversity & Evolution, p. 1–14, 2021.
HORTAL, J. et al. Seven shortfalls that beset large-scale knowledge of biodiversity. Annu. Rev.
Ecol. Evol. Syst., v. 46, p. 523–552, 2015.
HORTAL, J. et al. Accounting for biogeographical ignorance within biodiversity modelling.
Research Outreach, n. 129, 4 abr. 2022.
HUGHES, A. C. et al. The projected effects of climatic and vegetation changes on the
distribution and diversity of Southeast Asian bats. Global Change Biology, v. 18, n. 6, p. 1854–
1865, 2012.
JETZ, W.; MCPHERSON, J. M.; GURALNICK, R. P. Integrating biodiversity distribution
knowledge: Toward a global map of life. Trends in Ecology and Evolution, v. 27, n. 3, p. 151–
159, 2012.
KADMON, R.; FARBER, O.; DANIN, A. Effect of roadside bias on the accuracy of predictive
maps produced by bioclimatic models. Ecological Applications, v. 14, n. 2, p. 401–413, 2004.
KITCHENER, A. C. et al. A system for designating taxonomic certainty in mammals and other
taxa. Mammalian Biology, p. 1–11, 2022.
KLEPKA, V.; CORAZZA, M. J. O essencialismo na classificação de Lineu e a repercussão
dessa controvérsia na Biologia. História da Ciência e Ensino: construindo interfaces, v. 18, p.
73–110, 2018.
KONSTANTINOV, F. V.; NAMYATOVA, A. A. Taxonomic Revisions and Specimen
Databases in the Internet Age: Dealing with a Species Rich Insect Taxon. Entomological
Review, v. 99, n. 3, p. 340–361, 30 jun. 2019.
KUJALA, H.; MOILANEN, A.; GORDON, A. Spatial characteristics of species distributions as
drivers in conservation prioritization. Methods in Ecology and Evolution, v. 9, n. 4, p. 1121–
1132, 11 abr. 2018.
32

KUNIN, W. E. et al. Upscaling biodiversity: estimating the species–area relationship from small
samples. Ecological Monographs, v. 88, n. 2, p. 170–187, 2018.
LADLE, R.; HORTAL, J. Mapping species distributions: living with uncertainty. Frontiers of
Biogeography, v. 5, n. 1, 4 abr. 2013.
LADLE, R. J.; WHITTAKER, R. J. Conservation Biogeography. Oxford University Press,
Oxford., 2011.
LADLE, R. J.; WHITTAKER, R. J. Biogeografia e preservação ambiental. 1. ed. Oxford:
Wiley-Blackwell, 2014.
LESSA, T. et al. Known unknowns: Filling the gaps in scientific knowledge production in the
Caatinga. PLoS ONE, v. 14, n. 7, p. 1–12, 2019.
LESSA, T. et al. Quantifying spatial ignorance in the effort to collect terrestrial fauna in
Namibia, Africa. Ecological Indicators, v. 158, p. 111490, jan. 2024.
LESSA, T. et al. How taxonomic change influences forecasts of the Linnean shortfall (and what
we can do about it)? Journal of Biogeography, mar 2024.
LOBO, J. M. Database records as a surrogate for sampling effort provide higher species
richness estimations. Biodiversity and Conservation, v. 17, n. 4, p. 873–881, 2008.
LOMOLINO, M. V. Conservation biogeography. Frontiers of Biogeography: new directions in
the geography of nature, v. 293, 2004.
MAIR, L.; RUETE, A. Explaining Spatial Variation in the Recording Effort of Citizen Science
Data across Multiple Taxa. PLOS ONE, v. 11, n. 1, p. e0147796, 28 jan. 2016.
MAY, R. M. Tropical Arthropod Species, More or Less? Science, v. 329, n. 5987, p. 41–42, 2
jul. 2010.
MAYDEN, R. L. A hierarchy of species concepts: the denouement in the saga of the species
problem. In: Species: The units of diversity. [s.l.] Chapman & Hall, 1997. p. 381–423.
MAYR, E.; PROVINE, W. B. The Evolutionary Synthesis. [s.l.] Harvard University Press,
1980.
MEIJAARD, E.; NIJMAN, V. Primate Hotspots on Borneo: Predictive Value for General
Biodiversity and the Effects of Taxonomy. Conservation Biology, v. 17, n. 3, p. 725–732, 29
jun. 2003.
MEYER, C. et al. Global priorities for an effective information basis of biodiversity
distributions. Nature Communications, v. 6, n. 1, p. 8221, 8 set. 2015.
MORA, C. et al. How Many Species Are There on Earth and in the Ocean? PLoS Biology, v. 9,
n. 8, p. e1001127, 23 ago. 2011.
MOURA, M. R.; JETZ, W. Shortfalls and opportunities in terrestrial vertebrate species
discovery. Nature Ecology & Evolution, v. 5, n. 5, p. 631–639, 2021.
NABOUT, J. C. et al. The Impact of Global Climate Change on the Geographic Distribution
and Sustainable Harvest of Hancornia speciosa Gomes (Apocynaceae) in Brazil. Environmental
Management, v. 57, n. 4, p. 814–821, 2016.
33

NAVARRO-SIGÜENZA, A. G.; PETERSON, A. T. An alternative species taxonomy of the
birds of Mexico. Biota Neotropica, v. 4, n. 2, p. 1–32, 2004.
NEKOLA, J. C.; HORSÁK, M. The impact of empirically unverified taxonomic concepts on
ecological assemblage patterns across multiple spatial scales. Ecography, v. 2022, n. 5, p. 1–13,
2022.
NELSON, G.; ELLIS, S. The history and impact of digitization and digital data mobilization on
biodiversity research. Philosophical Transactions of the Royal Society B: Biological Sciences,
v. 374, n. 1763, p. 20170391, 7 jan. 2019.
OLIVEIRA, U. et al. The strong influence of collection bias on biodiversity knowledge
shortfalls of Brazilian terrestrial biodiversity. Diversity and Distributions, p. 1–13, 2016.
PADIAL, J. M. et al. The integrative future of taxonomy. Frontiers in Zoology, v. 7, n. 1, p. 16,
2010.
PANTE, E. et al. Species are hypotheses: avoid connectivity assessments based on pillars of
sand. Molecular Ecology, v. 24, n. 3, p. 525–544, 2015.
PARSONS, D. J. et al. Analysis of biodiversity data suggests that mammal species are hidden in
predictable places. Pnas, v. 119, n. 14, p. 1–7, 2022.
PETERSON, A. T.; NAVARRO-SIGÜENZA, A. G. Alternative species concepts as bases for
determining priority conservation areas. Conservation Biology, v. 13, p. 427–431, 1999.
RHEINDT, F. E. et al. Tightening the requirements for species diagnoses would help integrate
DNA-based descriptions in taxonomic practice. PLOS Biology, v. 21, n. 8, p. e3002251, 22 ago.
2023.
RIDDLE, B. R. et al. Basic Biogeography: Estimating Biodiversity and Mapping Nature. In:
Conservation Biogeography. [s.l: s.n.].
ROCCHINI, D. et al. Accounting for uncertainty when mapping species distributions: The need
for maps of ignorance. Progress in Physical Geography, v. 35, n. 2, p. 211–226, 2011.
ROCHA‐ORTEGA, M.; RODRIGUEZ, P.; CÓRDOBA‐AGUILAR, A. Geographical, temporal
and taxonomic biases in insect GBIF data on biodiversity and extinction. Ecological
Entomology, v. 46, n. 4, p. 718–728, 23 ago. 2021.
RODRIGUES, A. S. L. et al. The value of the IUCN Red List for conservation. Trends in
Ecology and Evolution, v. 21, n. 2, 2006.
ROSEN, R. On the limitations of scientific knowledge. In: CASTI, J.; KARLQVIST, A. (Eds.).
. oundaries and barriers: On the limits to scientific knowledge. [s.l.] MA: Perseus Books, 1996.
p. 199–214.
RUETE, A. Displaying bias in sampling effort of data accessed from biodiversity databases
using ignorance maps. Biodiversity Data Journal, v. 3, n. 1, 2015.
SANGSTER, G. Increasing numbers of bird species result from taxonomic progress, not
taxonomic inflation. Proceedings of the Royal Society B: Biological Sciences, v. 276, n. 1670,
p. 3185–3191, 2009.
SASTRE, P.; LOBO, J. M. Taxonomist survey biases and the unveiling of biodiversity patterns.
Biological Conservation, v. 142, n. 2, p. 462–467, fev. 2009.
34

SCOTLAND, R. W.; WORTLEY, A. H. How Many Species of Seed Plants Are There? Taxon,
v. 52, n. 1, p. 101–104, 2003.
SHEN, T. J.; CHAO, A.; LIN, C. F. Predicting the number of new species in further taxonomic
sampling. Ecology, v. 84, n. 3, p. 798–804, 2003.
SHIN, C. P.; ALLMON, W. D. How we study cryptic species and their biological implications:
A case study from marine shelled gastropods. Ecology and Evolution, v. 13, n. 9, 5 set. 2023.
SINGER, A. et al. Community dynamics under environmental change: How can next generation
mechanistic models improve projections of species distributions? Ecological Modelling, v. 326,
p. 63–74, abr. 2016.
SOUSA-BAENA, M. S.; GARCIA, L. C.; PETERSON, A. T. Completeness of digital
accessible knowledge of the plants of Brazil and priorities for survey and inventory. Diversity
and Distributions, v. 20, n. 4, p. 369–381, 2014.
STROPP, J. et al. Mapping ignorance: 300 years of collecting flowering plants in Africa. Global
Ecology and Biogeography, v. 25, n. 9, p. 1085–1096, 2016.
STROPP, J. et al. The ghosts of forests past and future: deforestation and botanical sampling in
the Brazilian Amazon. Ecography, v. 43, n. 7, p. 979–989, jul. 2020.
STROPP, J. et al. Taxonomic uncertainty and the challenge of estimating global species
richness. Journal of Biogeography, v. 49, n. 9, p. 1654–1656, 28 set. 2022.
TESSAROLO, G. et al. Temporal degradation of data limits biodiversity research. Ecology and
Evolution, v. 7, n. 17, p. 6863–6870, set. 2017.
TESSAROLO, G. et al. Using maps of biogeographical ignorance to reveal the uncertainty in
distributional data hidden in species distribution models. Ecography, v. 44, n. 12, p. 1743–1755,
2021.
THUILLER, W. et al. Predicting global change impacts on plant species’ distributions: Future
challenges. Perspectives in Plant Ecology, Evolution and Systematics, v. 9, n. 3–4, p. 137–152,
mar. 2008.
TROUDET, J. et al. Taxonomic bias in biodiversity data and societal preferences. Scientific
Reports, v. 7, n. 9132, p. 1–14, 2017.
VAN VALEN, L. Ecological species, multspecies, and Oaks. Taxon, v. 25, n. 2–3, p. 233–239,
26 maio 1976.
VEACH, V.; MOILANEN, A.; MININ, E. DI. Threats from urban expansion, agricultural
transformation and forest loss on global conservation priority areas. PLoS ONE, v. 12, n. 11, p.
1–14, 2017.
WHEELER, Q. A taxonomic renaissance in three acts. Megataxa, v. 1, n. 1, p. 4–8, 2020.
WHITTAKER, R. J. et al. Conservation Biogeography: Assessment and Prospect. Diversity and
Distributions, 2005.
WILEY, E. O. The Evolutionary Species Concept Reconsidered. Systematic Zoology, v. 27, n.
1, p. 17, mar. 1978.

35

ZACHOS, F. E. Species Concepts in Biology. Switzerland: Springer International Publishing,
2016.
ZACHOS, F. E. (New) Species concepts, species delimitation and the inherent limitations of
taxonomy. Journal of genetics, v. 97, n. 4, p. 811–815, 2018.
ZAPATA, F. A.; ROSS ROBERTSON, D. How many species of shore fishes are there in the
Tropical Eastern Pacific? Journal of Biogeography, v. 34, n. 1, p. 38–51, 2007.

36

Objetivos
Objetivo Geral
O objetivo geral desta tese foi avaliar, quantificar e discutir as incertezas e
ignorância nos dados disponíveis da biodiversidade sobre a taxonomia (lacuna
Linneana) e distribuição geográfica (lacuna Wallaceana).
Objetivos Específicos
Esta tese é dividida em três capítulos em formato de manuscrito, cujos objetivos
específicos foram:
Capítulo 1: How taxonomic change influences forecasts of the Linnean Shortfall
(and what we can do about it)
1. Discutir como a dinâmica do processo taxonômico afeta as estimativas de
espécies conhecidas e desconhecidas.
Capítulo 2: Do biological and ecological variables influence nomenclatural
uncertainty in birds?
1. Criar uma métrica para avaliar a incerteza nomenclatural das espécies de aves
globais.
2. Explorar as tendências da incerteza nomenclatural entre as espécies e Ordens de
aves globais.
3. Analisar associações entre variáveis biológicas e ecológicas das espécies de aves
globais e o escore de incerteza nomenclatural.
Capítulo 3: Quantifying spatial ignorance in the effort to collect terrestrial fauna in
Namibia, Africa
1. Aplicar a abordagem de escores de ignorância (“Ignorance Score”) para avaliar
as lacunas e vieses temporais, espaciais e taxonômicos nos registros de
ocorrência de espécies disponíveis no GBIF para Namíbia, África.
2. Analisar a influência das variáveis sociogeográficas na distribuição do esforço
de registros de ocorrência, a partir dos escores de ignorância.

37

Capítulo 1
Revista: Journal of Biogeography
Status: Aceito para publicação em 29/02/2024
How taxonomic change influences forecasts of the Linnean Shortfall
(and what we can do about it)
Running title: Taxonomic change and the Linnean Shortfall
Thainá Lessa1, Juliana Stropp2,3, Joaquín Hortal2,4, Richard J. Ladle1,5,6,*
Institutional Affiliation:
1

Institute of Biological and Health Sciences, Federal University of Alagoas, Maceió,

AL, Brazil
2

Department of Biogeography and Global Change, Museo Nacional de Ciencias

Naturales (MNCN-CSIC), Madrid, Spain
3

Department of Biogeography, Trier University, Trier, Germany.

4

Departamento de Ecologia, Instituto de Ciências Biológicas, Universidade Federal de

Goiás (UFG), Goiânia, Brazil
5

CIBIO, Centro de Investigação em Biodiversidade e Recursos Genéticos, InBIO

Laboratório Associado, Campus de Vairão, Universidade do Porto, 4485-661 Vairão,
Portugal
6

BIOPOLIS Program in Genomics, Biodiversity and Land Planning, CIBIO, Campus de

Vairão, 4485-661 Vairão, Portugal

*Corresponding author: Richard J. Ladle; CIBIO, Centro de Investigação em
Biodiversidade e Recursos Genéticos, InBIO Laboratório Associado, Campus de
Vairão, Universidade do Porto, 4485-661 Vairão, Portugal.
Email: richardjamesladle@gmail.com

38

Abstract
The gap between the number of described species and the number of species that
actually exist is known as the Linnean shortfall and is of fundamental importance for
biogeography and conservation. Unsurprisingly, there have been many attempts to
quantify its extent for different taxa and regions. In this Perspective we argue that such
forecasts remain highly problematic because the extent of the shortfall does not only
depend on the rates of exploration (sampling undescribed taxa) on which estimates have
been commonly based, but also on rates of taxonomic change (lumping and splitting).
These changes highly depend on the species concepts adopted and the information and
methods used to delimit species. Commonly used methods of estimating the number of
unknown species (e.g., discovery curves, taxon ratios) can underestimate or
overestimate the Linnean shortfall if they don’t effectively account for trends and rates
of taxonomic change. A further complication is that the history of taxonomic change is
not well documented for most taxa and is not typically available in biodiversity
databases. Moreover, wide geographic and taxonomic variation in the adoption of
species concepts and delimitation methods mean that comparison of estimates of the
Linnean shortfall between taxa and even for the same taxon between regions may be
unreliable. Given the high likelihood of future taxonomic changes for most major taxa,
we propose two main strategies to consider the influence of taxonomic change on
estimates of unknown species; i) a highly conservative approach to estimating the
Linnean shortfall, restricting analysis to groups and regions where taxonomies are
relatively stable; and ii) explicitly incorporating metrics of taxonomic change into
biodiversity models and estimates. In short, relevant estimates of the number of known
and unknown species will only be achieved by accounting for the dynamic nature of the
taxonomic process itself.
Keywords: species descriptions, taxonomic fluctuations, taxonomic lumping and
splitting, global number of species, unknown species, shortfalls.

39

Introduction
Species are arguably the most important units for measuring biodiversity, and rates of
species loss (extirpation and extinction) are key statistics for tracking human impacts on
the environment (Ladle & Whittaker, 2011). Nevertheless, over 250 years after Carl von
Linné (1707-1778) began his systematic inventory of the World’s species, there is still
considerable debate about how many species actually exist and how many of these we
have already documented. The difference between the former and the latter is known as
the Linnean shortfall (Brown & Lomolino, 1998), and is of fundamental importance for
ecology, biogeography and the conservation of the Earth’s remaining biological
resources (Hortal et al., 2015; Whittaker et al., 2005). For example, unrecognized
variation between regions in the proportion of unknown species could lead to
misidentification of biodiversity patterns with knock-on effects for conservation
prioritization and the inference of ecological and evolutionary processes (Diniz Filho et
al., 2023; Edie et al., 2017; Stropp et al., 2022).
The challenge for biogeographers is that the numbers of known species are
affected by the discovery of new species and the taxonomic reorganization of already
known taxa. This leads to different types of unknown species. First, those species that
are yet to be sampled. These are probably most common in the few remaining large,
under-surveyed regions of the World such as the tropical moist broadleaf forests (Moura
& Jetz, 2021), forests of southwest Amazonia (Hopkins, 2007, 2019) or the Brazilian
Caatinga dry forest (Lessa et al., 2019), and poorly studied ecosystems such as the deep
sea (Danovaro et al., 2010) or the upper canopies of rainforests (Ellwood & Foster,
2004). The number of these ‘yet to be sampled species’ is probably quite small for wellknown groups of vertebrates, and much larger for less well-sampled taxa such as many
invertebrates (Cardoso et al. 2011) and many largely unknown microbial groups.
Second, those species that have already been sampled but are yet to be formally
described. These include historical specimens in museums and other natural history
collections that have never been properly evaluated (Bebber et al., 2010). The number
of such species may conceivably run into the hundreds of thousands, many of which
may already be extinct (Alberch, 1993). In addition, there are unknown species that are
currently “hidden” within another species, but which will be upgraded to accepted
species status (i.e. taxonomic splitting) after a taxonomic re-evaluation (see Parsons et
al., 2022). Finally, there are many currently valid (accepted) species that may be

40

aggregated into a single species (i.e. taxonomic lumping) in the future due to taxonomic
revision.
These different types of unknown species and taxonomic reorganizations have
direct consequences for the estimation of the Linnean shortfall. Specifically, estimates
of unknown species typically only account for new discoveries, thus failing to account
for the fluctuations in species numbers that emerge purely from taxonomic
reorganizations, such as new species created by splitting a taxon into two or more
during taxonomic revision, or the synonymization of two or more taxa which are
lumped into a single one. This is not a trivial problem, since changes in species
designations are inherent to the taxonomic process (Hobern et al., 2021; Thiele et al.,
2021) and may significantly outstrip new species discoveries in many taxa (Simkins et
al., 2020). In this Perspective we discuss how we might account for the dynamics
associated with the taxonomic process to improve our estimates of known and unknown
species.
Extrapolating from uneven foundations
Many methods have been proposed to measure the Linnean shortfall. Most of these
methods, even those based on expert opinion, are ultimately linked to counts of
currently valid species, be that extrapolations of species discovery trends or from wellknown taxa or intensively studied areas. In an ideal world, each newly described species
would be meticulously documented, unambiguously identified, named, and definitively
and permanently allocated a unique branch on the tree of life using identical methods
based on a single, universally applied concept of what constitutes a species (Stropp et
al, 2022). In the real world, taxonomy is built on an uneven foundation of different
species concepts and delimitation methods (Zachos, 2016, 2018a). Furthermore, the
rules established by the International Codes of Nomenclature allow freedom in the
choice of criteria for species comparison and diagnoses, which increase the chances of
new taxonomic changes and, concomitantly, decrease taxonomic stability (Rheindt et al.
2023).
The application of different species concepts and delimitation methods
inevitably leads to multiple propositions for the number of species within a taxon or
geographic region – even for well-known taxa such as birds, multiple global species
lists are still in operation (Neate-Clegg et al., 2021). The difficulties associated with
producing a single, universally accepted list of species have recently been discussed in a
41

series of articles (Conix et al., 2021; Hobern et al., 2021; Lien et al., 2021; Pyle et al.,
2021; Thiele et al., 2021; Thomson et al., 2021). Specifically, they highlight that for any
such list to be widely adopted, mechanisms would have to be developed to ensure: i) the
accuracy and consistency of additions to the list; ii) that the list is regularly updated and
maintained; iii) that obscure and newly described taxa are consistently dealt with, and;
iv) there are robust mechanisms for arbitrating (the inevitable) disputes or alternative
taxonomic viewpoints. Details concerning how a global checklist can be accessed, how
it will be maintained, and the way in which the list and its contents are properly cited
still need to be determined. Some of these issues have been considered, and sometimes
solved, by the Catalogue of Life (Bánki et al., 2023), which included 2,121,194 species
as of 28th February 2024 (https://www.catalogueoflife.org).
Taxonomy is a dynamic discipline with great variation in practices over time, in
different geographic regions and often between taxonomists working on different (or
even the same) groups of organisms (Nekola & Horsák, 2022; Pyle et al., 2021). For
example, arthropods are known to have the majority undescribed species, either due to
incipient exploration or because there are species hidden in others (Costello et al.,
2013). However, even for well-known groups, such as mammals and birds, these gaps
are not completely filled (Mora et al., 2011; Parsons et al., 2022). In short, the Linnean
shortfall is generally more severe for organisms that are smaller in size, niche width, or
distributional range and which are less complex or phenotypically conspicuous, with
this pattern holding both between and within taxonomic groups (Zapata & Robertson,
2006; Riddle et al., 2011; dos Santos et al., 2020; Guedes et al., 2023; Poulin et al.,
2023). Recent research suggests that scientists have still only robustly sampled 6.74%
of the Earth, with the tropics, high elevations and deep seas especially poorly covered
(Hughes et al., 2021). It is possible that in the future we will achieve a high degree of
taxonomic synthesis for some taxa (Nekola & Horsák, 2022), but we are not there yet.
Alternatively, we may adopt a different way of accounting for nature such as the
recently proposed multilevel organismal diversity concept (Martynov & Korshunova,
2022), the widespread adoption of which has the potential to radically change biological
nomenclature and perceptions of biodiversity patterns.
Lumping versus splitting: a zero-sum game?
The number of known species can either increase or decrease, even in the absence of
new discoveries (Figure 1). Increases occur when existing subspecies/races/populations
42

are raised to species level, ‘splitting’ a formerly-recognized species into two or more
due to changes in the application of species concepts or species delimitation methods,
especially in recently described species delimited based on genetic information (Isaac et
al., 2004). Such increases are frequently a sign of taxonomic progress (Korshunova et
al., 2023) and have been identified in many taxa, but are particularly prevalent in wellknown vertebrate groups such as birds (Simkins et al., 2020) and mammals (Gippoliti &
Groves, 2012; Zachos, 2018b). For example, a recent analysis of temporal trends in
known mammal species numbers found an increase of 1,079 species over 13 years,
nearly all due to taxonomic revisions (Burgin et al., 2018), a trend that is forecasted to
continue (Parsons et al., 2022). Moreover, taxonomic changes are ongoing and it is
highly probable that, for example, many intraspecific taxa (e.g., subspecies) of birds
will be formally recognized as valid species over the forthcoming decades
(Barrowclough et al., 2016). There have been intense debates about whether such
increases (often termed ‘taxonomic inflation’) are justified (Garnett & Christidis, 2017;
Harris & Froufe, 2005; Padial & De la Riva, 2006; Sangster, 2009). From the
perspective of the Linnean shortfall, high frequencies of future taxonomic splits will
lead to current underestimates of the total number of species in a taxon using standard
forecasting methods such as extrapolation of discovery curves (Figure 1).
The number and proportion of species that will be created by taxonomic splits,
as opposed to undiscovered species, is difficult to evaluate for all but the best known
taxa (e.g., Parsons et al., 2022). For birds, estimates of the number of species in the
world remained remarkably consistent over the second half of the 20th century. In 1946,
Ernst Mayr, the first proponent of the biological species concept, estimated there to be
approximately 8,600 species (Mayr, 1946). This figure had only slightly changed by the
end of the century; Sibley & Monroe (1990) suggested there may be as many as 9,700
bird species. With the introduction of more sophisticated molecular techniques and
widespread changes in taxonomic practice, these estimates have continued to increase
along with the number of officially recognized species (notwithstanding that there are
multiple global checklists). For example, the current list of the International
Ornithological Committee (IOC) recognizes 10,928 extant species (Gill et al., 2022),
whereas the list given by the Handbook of the Birds of the World (HBW) recognizes
10,824 extant species (del Hoyo et al., 2013). It is an ongoing question of how many of
these species might be revised (either split or lumped, or both) in the future. When a
diagnostic, evolutionary species concept was applied to a morphological and
43

distributional data set from 200 species, it was estimated that there could be as many as
18,043 species of birds worldwide (95% confidence interval of 15,845 to 20,470;
Barrowclough et al., 2016). This figure is close to the current number (19,883) of
recognized sub-species on the IOC list. How many of these sub-species will eventually
be upgraded to full species status? This is almost impossible to answer, since it depends
on continued changes in taxonomic theory and practice. Whatever the final figure is,
recent history (Simkins et al. 2020) suggests it will be significantly greater than the
proportion of newly discovered bird species in the field.

Figure 1: Schematic representation of the impact of lumping and splitting on estimates of the
number of known and unknown species (Linnean Shortfall). The centre of the graph represents
taxonomic stability (no further splitting or lumping of species). A scenario of more splitting
than lumping leads to underestimates of the shortfall (right lower quadrant, blue), while more
lumping than splitting leads to overestimates of the shortfall (left upper quadrant, yellow). For
example, there was relatively more splitting than lumping of mammals between 2005 and 2017
(figures from Burgin et al. 2018), meaning pre-2005 estimates of the total number of mammal
species (unknown + known) were almost certainly underestimates. Size of animal silhouettes is
proportional to the number of valid species, with grey indicating pre-2005 and black post-2005.
44

The taxonomic history of birds and mammals is not necessarily reflective of
other taxa (Knapp et al., 2005), and taxonomic revisions can also decrease the number
of valid species. A proportion of currently recognized species will, at some point in the
future, be relegated to synonyms due to ‘lumping’. These lumping events are most
common in early described species where its delimitation was based on morphological
traits. Bouchet (2006) suggested that as many as a fifth of all recently described species
names may become synonyms. In the case of amphibians, Hillis (2019) predicts that the
future is likely to see high levels of lumping as more careful analysis corrects
widespread ‘over-splitting’ that occurred when taxonomists began to introduce
molecular methods and adopt the phylogenetic species concept.
In some taxa, such as Neotropical palms, the number of synonyms can outstrip
the number of valid/accepted species (Stropp et al., 2022). This way, taxonomic change
leads to a reduction in the number of valid/accepted species. For example, the 27
species that currently comprise the palm genus Attalea (Henderson et al., 1995) have
previously been described as 16 different genera and associated with 167 species at
different points over the past two hundred years (Henderson, 2020). The most recent
revision of the Attalea proposes 30 species, which are associated with 93 heterotypic
synonyms (data from Henderson 2020). It is important to remember that heterotypic
synonyms produced by taxonomic lumping were once considered as valid species and
consequently, if the Linnean shortfall had been calculated before heterotypic
synonymizations were proposed, the scientists of the time would have vastly
overestimated the number of unknown species using forecasting methods such as
extrapolating from discovery curves (e.g., Bebber et al., 2007; Stropp et al., 2022).
There is no reason to assume that, at the level of higher taxa, new species created
by splitting will be broadly ‘compensated’ by losses due to lumping. Rather, historical,
current and future levels of splitting and lumping will vary according to biocultural
factors, including the number of taxonomists working on a particular taxon in a
particular region (Freeman & Pennell, 2021), levels of cryptic biodiversity (Beheregaray
& Caccone, 2007), geographical biases in taxonomic practices and the completeness of
biological collections (Harris & Froufe, 2005; Lavoie, 2013), among others. In other
words, splitting and lumping cannot be assumed to be a ‘zero-sum game’ and
taxonomic trends may vary enormously even between closely related taxa (Williams,
2022). As we include more diverse taxa in our analysis, systematic biases between
taxonomic groups caused by splitting and lumping practices may become even more
45

pronounced because of different states of taxonomic knowledge and continued progress
in taxonomy (Lughadha et al., 2016; Troudet et al., 2017).
Is it possible to account for the taxonomic changes?
There are eleven general approaches for estimating the number of undiscovered species,
each with its own assumptions. It is certainly possible to make broad predictions about
how many species remain unknown (either undiscovered or unrecognized) based on
these approaches. However, in our opinion there is a largely unrecognized caveat that
future taxonomic revisions have the potential to significantly increase or decrease
forecasts. Even if this caveat is widely known, the vast majority of published forecasts
for the Linnean shortfall fail to account for, or simply ignore, the impacts of taxonomic
change on their estimates (e.g., Costello et al., 2015; Costello & Wilson, 2011; Gatti et
al., 2022; Legros et al., 2020; Moura & Jetz, 2021).
Dealing with the effects of taxonomic change on forecasts of unknown species is
by no means simple, but this does not justify continuing to ignore this ‘elephant in the
room’ of biogeographical and macroecological research. In our opinion there are two
main approaches that can be adopted to deal with taxonomic change, though both have
significant limitations. Firstly, we could be cautious and only make extrapolations for
groups with a fairly stable taxonomy or within geographic areas with uniform
taxonomic practice (Freeman & Pennell, 2021). Smaller taxonomic and geographic
scales would certainly improve forecasts and could potentially be aggregated.
Nevertheless, as taxonomy is currently undergoing a period of intense change in
practices (Wheeler, 2020).
A second approach is to better account for taxonomic change in models and
forecasts. This could potentially be achieved by identifying and quantifying how
lumping, splitting, or new discoveries have influenced past biodiversity estimates.
Unfortunately, gathering data on historical taxonomic change is difficult and very
labour intensive because most, taxonomic databases only provide accounts of currently
accepted and not accepted names, and do not give the full history of changes associated
with these names (Franz & Peet, 2010). It would be useful to know for how long and
when each name was associated with a valid species. One potential barrier to obtaining
historical taxonomic information is the current difficulty of accessing and collecting
relevant data for a wide range of taxa. Data on when names cease to be valid are often
scattered in the taxonomic literature and not easily accessible via the generally-available
46

databases and checklists. Indeed, most taxonomic databases typically codify only when
a name was proposed and/or reinstated. Some notable examples could be used as
references for other taxa, such as Amphibian Species of the World, The reptile database
and International Ornithological Community, organizations that collate and share
taxonomic information for amphibians, reptiles and birds, respectively. However, this is
further complicated by the fact that species are hypotheses (Pante et al., 2015), and
proposals for changes in taxonomic status are not necessarily accepted or adopted by the
taxonomic community. Furthermore, acceptance of a proposed change does not occur
immediately. A partial solution to this could be to use culturomics analysis (Ladle et al.,
2016). Here, analyses of changes in name frequency in the academic literature could
help identify when species ‘hypotheses’ were adopted or rejected, though a formal
method to achieve this has not yet been proposed (but see Newberry & Plotkin, 2022).
Taxonomic databases such as the Catalogue of Life (Bánki et al., 2023) provide
information on synonyms, but not when and for how long these were considered as
valid species. The reality is that reconstructing the history of taxonomic change often
requires painstaking research on the taxonomic literature proposing heterotypic
synonyms (Creighton et al., 2022; Vaidya et al., 2018). This makes analysis of historical
taxonomic trends for a large number of taxa or large geographic regions highly
challenging, especially for non-taxonomists (such as many biogeographers and
macroecologists). Advances in text mining technologies based on machine learning
(Farrell et al., 2022) and the widespread digitalization of taxonomic monographs will
certainly help reduce this data gap. Once these data are available we could model the
probability of taxonomic change associated with each accepted species name. In turn,
the results of these models can be used to identify taxon - or geographically- specific
scenarios of taxonomic change and to incorporate these into forecasting models (Alroy,
2002; Edie et al., 2017).
Conclusions: Revisiting the Linnean Shortfall
The discrepancy between the number of species recognized to exist now and how many
actually exist (if we had access to all the relevant data) is highly dependent on how we
define and evaluate species. Moreover, taxonomic tools and practices are rapidly
evolving and are likely to do so for many years to come (Gill, 2014; Padial & De la
Riva, 2021; Wheeler, 2020). Much of these changes are due to advances in molecular
and computational methods, combined with a large number of competing species
47

concepts jostling for dominance (Kitchener et al., 2022; Stankowski & Ravinet, 2021;
Zachos, 2018b). The culture of taxonomy may also be changing, with a greater
willingness among taxonomists to designate new species from allopatric or parapatric
populations that were previously part of polytypic species (Meiri & Mace, 2007). The
discrepancies in taxonomic practices (Harris & Froufe, 2005; Stankowski & Ravinet,
2021), coupled with uneven biological knowledge for different taxa and regions (e.g.,
Meyer et al., 2016), mean that estimates of the Linnean shortfall at scale are, and will
remain, immensely challenging.
Biogeographers and macroecologists need to be aware that the Linnean shortfall
is a biocultural phenomenon whose characteristics depend on the complex interplay of
various factors, including: i) history of where we have explored; ii) how much effort we
have expended searching for different taxa; iii) how easy or difficult a species is to
observe/collect; iv) how we define and delimit species; v) which is the level of
digitalization and data mobilisation of the taxonomic work already devoted to each taxa;
vi) how long and when each name is associated with an accepted species; and vii) how
we use the data on how many species exist within a defined area to predict how many
species remain undiscovered in that area. Both the numbers of known and unknown
species are therefore subject to considerable fluctuations due to changes in taxonomic
practice and, with them, to a significant degree of taxonomic uncertainty, thus calling
for considerable caution when conducting and interpreting global estimates. This caveat
may be addressed by designing strategies to account for taxonomic change when
estimating current diversity or for forecasting the numbers of unknown species. These
new estimates should account for the species that are yet to be discovered in the field,
those that one day will be split from currently recognized taxa, and those that will be
aggregated due to taxonomic lumping. Only when the resulting fluctuations in species
number stemming from these three processes are integrated into the biogeographers’
analytical toolbox will our models of global biodiversity patterns and processes become
truly robust. This necessity is increasingly being acknowledged by the ecology and
biogeography community, as evidenced by recent proposals to tighten requirements for
species comparisons and diagnoses (Rheindt et al. 2023) and to create a globally
integrated structure for taxonomy (Sandall et al. 2023). Such advances need necessarily
to go a long way towards creating a more stable taxonomy that provides a solid basis for
extrapolations of the Linnean Shortfall.

48

Acknowledgements: This work was funded by the European Union’s Horizon 2020
research and innovation programme (grant agreement #854248). JS was also supported
by EU H2020 Marie Skłodowska-Curie Action TAXON-TIME (grant agreement
#843234), and JH by project SCENIC (grant PID2019-106840GB-C21, funded by
Spanish MCIN/AEI/ 10.13039/501100011033. RJL is supported by the European
Union’s Horizon 2020 research and innovation programme under grant agreement no.
854248. TL was funded by the “Coordenação de Aperfeiçoamento de Pessoal de Nível
Superior” - Brazil (CAPES) - Finance Code 001. Additional support has been provided
by the “Instituto Nacional de Ciência e Tecnologia em Ecologia, Evolução e
Conservação da Biodiversidade” (INCT-EECBio) (MCTIC/CNPq 465610/2014-5;
FAPEG 201810267000023). No fieldwork was conducted so no collection permits were
required for this work.
References
Alberch, P. (1993). Museums, collections and biodiversity inventories. Trends in Ecology &
Evolution, 8(10), 372–375.
Alroy, J. (2002). How many named species are valid? Pnas, 99(6), 3706–3711.
https://doi.org/https://doi.org/10.1073/pnas.062691099
Bánki, O., Roskov, Y., Döring, M., Ower, G., Hernández Robles, D. R., Plata Corredor, C. A.,
Stjernegaard Jeppesen, T., Örn, A., Vandepitte, L., Hobern, D., Schalk, P., DeWalt, R. E., Ma,
K., Miller, J., Orrell, T., Aalbu, R., Abbott, J., Adlard, R., Adriaenssens, E. M., et al. (2023).
Catalogue
of
Life
Checklist
(Version
2023-08-17).
Catalogue
of
Life.
https://doi.org/10.48580/dft7
Barrowclough, G. F., Cracraft, J., Klicka, J., & Zink, R. M. (2016). How many kinds of birds
are there and why does it matter? PLoS One, 11(11), e0166307.
Bebber, D. P., Carine, M. A., Wood, J. R. I., Wortley, A. H., Harris, D. J., Prance, G. T.,
Davidse, G., Paige, J., Pennington, T. D., Robson, N. K. B., & Scotland, R. W. (2010). Herbaria
are a major frontier for species discovery. Pnas, 107(51), 22169–22171.
https://doi.org/10.1073/pnas.1011841108
Bebber, D. P., Marriott, F. H. C., Gaston, K. J., Harris, S. A., & Scotland, R. W. (2007).
Predicting unknown species numbers using discovery curves. Proceedings of the Royal Society
B: Biological Sciences, 274(1618), 1651–1658.
Beheregaray, L. B., & Caccone, A. (2007). Cryptic biodiversity in a changing world. Journal of
Biology, 6(4), 1–5.
Bouchet, P. (2006). The magnitude of marine biodiversity. The Exploration of Marine
Biodiversity: Scientific and Technological Challenges, 31–62.
Brown, J. H., & Lomolino, M. V. (1998). Biogeography.

49

Burgin, C. J., Colella, J. P., Kahn, P. L., & Upham, N. S. (2018). How many species of
mammals are there? Journal of Mammalogy, 99(1), 1–14.
Conix, S., Garnett, S. T., Thiele, K. R., Christidis, L., van Dijk, P. P., Bánki, O. S., Barik, S. K.,
Buckeridge, J. S., Costello, M. J., & Hobern, D. (2021). Towards a global list of accepted
species III. Independence and stakeholder inclusion. Organisms Diversity & Evolution, 21(4),
631–643.
Costello, M.J., May, R. M., Stork, N. E. (2013). Can we name Earth’s species before they go
extinct? Science, 339, 413–416. https://doi.org/10.1126/science.1230318
Costello, M. J., Lane, M., Wilson, S., & Houlding, B. (2015). Factors influencing when species
are first named and estimating global species richness. Global Ecology and Conservation, 4,
243–254. https://doi.org/10.1016/j.gecco.2015.07.001
Costello, M. J., & Wilson, S. P. (2011). Predicting the number of known and unknown species
in European seas using. Global Ecology and Biogeography, 20, 319–330.
https://doi.org/10.1111/j.1466-8238.2010.00603.x
Creighton, M. J. A., Luo, A. Q., Reader, S. M., & Mooers, A. Ø. (2022). Predictors of
taxonomic splitting and its role in primate conservation. Animal Conservation.
https://doi.org/https://doi.org/10.1111/acv.12825
Danovaro, R., Company, J. B., Corinaldesi, C., D’Onghia, G., Galil, B., Gambi, C., Gooday, A.
J., Lampadariou, N., Luna, G. M., & Morigi, C. (2010). Deep-sea biodiversity in the
Mediterranean Sea: the known, the unknown, and the unknowable. PloS One, 5(8), e11832.
del Hoyo, J., Elliott, A., Sargatal, J., Christie, D. A., del Hoyo, J., Elliott, A., & Sargatal, J.
(2013). Handbook of the birds of the world. Lynx edicions.
Diniz Filho, J. A. F., Jardim, L., Guedes, J. J. M., Meyer, L., Stropp, J., Frateles, L. E. F., Pinto,
R. B., Lohmann, L. G., Tessarolo, G., & de Carvalho, C. J. B. (2023). Macroecological links
between the Linnean, Wallacean, and Darwinian shortfalls. Frontiers of Biogeography.
dos Santos, J.W., Correia, R.A., Malhado, A.C.M., Campos-Silva, J.V., Teles, D., Jepson, P.,
Ladle, R.J. (2020). Drivers of taxonomic bias in conservation research: a global analysis of
terrestrial mammals. Anim. Conserv., 23, 679-688. https://doi.org/10.1111/acv.12586
Edie, S. M., Smits, P. D., & Jablonski, D. (2017). Probabilistic models of species discovery and
biodiversity comparisons. Pnas, 114(14), 3666–3671. https://doi.org/10.1073/pnas.1616355114
Ellwood, M. D. F., & Foster, W. A. (2004). Doubling the estimate of invertebrate biomass in a
rainforest canopy. Nature, 429(6991), 549–551.
Farrell, M. J., Brierley, L., Willoughby, A., Yates, A., & Mideo, and N. (2022). Past and future
uses of text mining in ecology and evolution. Proceedings of the Royal Society B, 289(1975),
1–9. https://doi.org/https://doi.org/10.1098/rspb.2021.2721
Franz, N. M., & Peet, R. K. (2010). Perspectives: Towards a language for mapping relationships
among
taxonomic
concepts.
Systematics
and
Biodiversity,
7(1),
5–20.
https://doi.org/https://doi.org/10.1017/S147720000800282X
Freeman, B. G., & Pennell, M. W. (2021). The latitudinal taxonomy gradient. Trends in
Ecology & Evolution, 1, 1–9. https://doi.org/10.1016/j.tree.2021.05.003

50

Garnett, S. T., & Christidis, L. (2017). Taxonomy anarchy hampers conservation. Nature News,
546(7656), 25.
Gatti, R. C., Reich, P. B., Gamarra, J. G. P., Crowther, T., Hui, C., Morera, A., Bastin, J.-F., deMiguel, S., Nabuurs, G.-J., & Svenning, J.-C. (2022). The number of tree species on Earth.
Proceedings of the National Academy of Sciences, 119(6).
Gill,
Donsker,
&
Rasmussen.
https://www.worldbirdnames.org/new/

(2022).

IOC

World

Bird

List.

Gill, F. B. (2014). Species taxonomy of birds: which null hypothesis? The Auk: Ornithological
Advances, 131(2), 150–161.
Gippoliti, S., & Groves, C. P. (2012). “Taxonomic inflation” in the historical context of
mammalogy and conservation. Hystrix, 23(2), 8.
Guedes, J. J. M., Moura, M. R., Diniz-Filho, J. A. F. (2023). Species out of sight: elucidating
the determinants of research effort in global reptiles. Ecography. e06491.
https://doi.org/10.1111/ecog.06491
Harris, D. J., & Froufe, E. (2005). Taxonomic inflation: species concept or historical
geopolitical bias? Trends in Ecology & Evolution, 20(1), 6–7.
Henderson, A. J. (2020). A revision of Attalea (Arecaceae, Arecoideae, Cocoseae, Attaleinae).
Phytotaxa, 444, 1–76. https://doi.org/10.11646/phytotaxa.444.1.1
Henderson, A. J., Galeano, G., & Bernal, R. (1995). A field guide to the palms of the Americas.
Princeton University Press. https://doi.org/10.2307/j.ctvcszzzd
Hillis, D. M. (2019). Species delimitation in herpetology. Journal of Herpetology, 53(1), 3–12.
Hobern, D., Barik, S. K., Christidis, L., T Garnett, S., Kirk, P., Orrell, T. M., Pape, T., Pyle, R.
L., Thiele, K. R., & Zachos, F. E. (2021). Towards a global list of accepted species VI: The
Catalogue of Life checklist. Organisms Diversity & Evolution, 1–14.
Hopkins, M. J. G. (2007). Modelling the known and unknown plant biodiversity of the Amazon
Basin. Journal of Biogeography, 34, 1400–1411. https://doi.org/10.1111/j.13652699.2007.01737.x
Hopkins, M. J. G. (2019). Are we close to knowing the plant diversity of the Amazon? Anais
Da Academia Brasileira de Ciências, 91.
Hortal, J., Bello, F. de, Diniz-Filho, J. A. F., Lewinsohn, T. M., Lobo, J. M., & Ladle, R. J.
(2015). Seven shortfalls that beset large-scale knowledge of biodiversity. Annu. Rev. Ecol.
Evol. Syst., 46, 523–552. https://doi.org/10.1146/annurev-ecolsys-112414-054400
Hughes, A. C., Orr, M. C., Ma, K., Costello, M. J., Waller, J., Provoost, P., Yang, Q., Zhu, C.,
& Qiao, H. (2021). Sampling biases shape our view of the natural world. Ecography.
Isaac, N. J. B., Mallet, J., & Mace, G. M. (2004). Taxonomic inflation: its influence on
macroecology and conservation. Trends in Ecology & Evolution, 19(9), 464–469.
Kitchener, A. C., Hoffmann, M., Yamaguchi, N., Breitenmoser-Würsten, C., & Wilting, A.
(2022). A system for designating taxonomic certainty in mammals and other taxa. Mammalian
Biology, 1–11.
51

Knapp, S., Lughadha, E. N. N., & Paton, A. (2005). Taxonomic inflation, species concepts and
global
species
lists.
Trends
in
Ecology
&
Evolution,
1(20),
7–8.
https://doi.org/10.1016/j.tree.2004.10.012
Korshunova, T., Lundin, K., Malmberg, K., & Martynov, A. (2023). Narrowly defined taxa on a
global scale: The phylogeny and taxonomy of the genera Catriona and Tenellia (Nudibranchia,
Trinchesiidae) favours fine‐scale taxonomic differentiation and dissolution of the “lumpers &
splitters” dilemma. Evolutionary Applications, 16(2), 428-460.
Ladle, R. J., Correia, R. A., Do, Y., Joo, G., Roberge, J., & Jepson, P. (2016). Conservation
culturomics.
Frontiers
in
Ecology
and
the
Environment,
14,
269–275.
https://doi.org/10.1002/fee.1260
Ladle, R. J., & Whittaker, R. J. (2011). Conservation Biogeography. In Conservation
Biogeography. https://doi.org/10.1002/9781444390001
Legros, V., Rochat, J., Reynaud, B., & Strasberg, D. (2020). Known and unknown terrestrial
arthropod fauna of La Réunion Island, Indian Ocean. Journal of Insect Conservation, 24, 199–
217.
Lessa, T., Dos Santos, J. W., Correia, R. A., Ladle, R. J., & Malhado, A. C. M. (2019). Known
unknowns: Filling the gaps in scientific knowledge production in the Caatinga. PLoS ONE,
14(7), 1–12. https://doi.org/10.1371/journal.pone.0219359
Lien, A. M., Conix, S., Zachos, F. E., Christidis, L., van Dijk, P. P., Bánki, O. S., Barik, S. K.,
Buckeridge, J. S., Costello, M. J., & Hobern, D. (2021). Towards a global list of accepted
species IV: Overcoming fragmentation in the governance of taxonomic lists. Organisms
Diversity & Evolution, 21(4), 645–655.
Lughadha, E. N., Govaerts, R., Belyaeva, I., Black, N., Lindon, H., Allkin, R., Magill, R. E., &
Nicolson, N. (2016). Counting counts: revised estimates of numbers of accepted species of
flowering plants, seed plants, vascular plants and land plants with a review of other recent
estimates. Phytotaxa, 272(1), 82. https://doi.org/10.11646/phytotaxa.272.1.5
Martynov, A.V. & Korshunova, T.A. (2022). Multilevel Organismal Diversity in an ontogenetic
framework as a solution for the species concept. In: Cryptic Species Morphological Stasis,
Circumscription, and Hidden Diversity. Cambridge University Press. p. 78–129.
Mayr, E. (1946). The number of species of birds. The Auk, 63(1), 64–69.
Meiri, S., & Mace, G. M. (2007). New taxonomy and the origin of species. Public Library of
Science San Francisco, USA.
Meyer, C., Jetz, W., Guralnick, R. P., Fritz, S. A., & Kreft, H. (2016). Range geometry and
socio‐economics dominate species‐level biases in occurrence information. Global Ecology and
Biogeography, 25(10), 1181–1193.
Mora, C., Tittensor, D. P., Adl, S., Simpson, A. G. B., & Worm, B. (2011). How Many Species
Are There on Earth and in the Ocean? PLoS Biology, 9(8), e1001127.
https://doi.org/10.1371/journal.pbio.1001127
Moura, M. R. R., & Jetz, W. (2020). Shortfalls and opportunities in terrestrial vertebrate species
discovery. BioRxiv, 5(5), 631–639. https://doi.org/10.1038/s41559-021-01411-5

52

Neate-Clegg, M. H. C., Blount, J. D., & Şekercioğlu, Ç. H. (2021). Ecological and
biogeographical predictors of taxonomic discord across the world’s birds. Global Ecology and
Biogeography, 30(6), 1258–1270. https://doi.org/10.1111/geb.13300
Nekola, J. C., & Horsák, M. (2022). The impact of empirically unverified taxonomic concepts
on ecological assemblage patterns across multiple spatial scales. Ecography, 2022(5), 1–13.
Newberry, M. G., & Plotkin, J. B. (2022). Measuring frequency-dependent selection in culture.
Nature Human Behaviour, 6, 1048–1055. https://doi.org/s41562-022-01342-6
Padial, J. M., & De la Riva, I. (2006). Taxonomic inflation and the stability of species lists: the
perils of ostrich’s behavior. Systematic Biology, 55(5), 859–867.
Padial, J. M., & De la Riva, I. (2021). A paradigm shift in our view of species drives current
trends in biological classification. Biological Reviews, 96(2), 731–751.
Pante, E., Puillandre, N., Viricel, A., Arnaud-Haond, S., Aurelle, D., Castelin, M., Chenuil, A.,
Destombe, C., Forcioli, D., Valero, M., Viard, F., Samadi, S., Arnaud‐Haond, S., Aurelle, D.,
Castelin, M., Chenuil, A., Destombe, C., Forcioli, D., & Valero, M. (2015). Species are
hypotheses: avoid connectivity assessments based on pillars of sand. Molecular Ecology, 24(3),
525–544. https://doi.org/https://doi.org/10.1111/mec.13048
Parsons, D. J. J., Pelletier, T. A., Wieringa, J. G. G., Duckett, D. J., Carstens, B. C. C.,
Pelletierc, T. A., Wieringa, J. G. G., Ducketta, D. J., & Carstens, B. C. C. (2022). Analysis of
biodiversity data suggests that mammal species are hidden in predictable places. Proceedings of
the
National
Academy
of
Sciences,
119(14),
e2103400119.
https://doi.org/10.1073/pnas.2103400119
Poulin R., Presswell B., Bennett J., Dutra, D. A., Salloum, P. M. (2023). Biases in parasite
biodiversity research: why some helminth species attract more research than others,
International Journal for Parasitology: Parasites and Wildlife, 21, 89-98,
https://doi.org/10.1016/j.ijppaw.2023.04.010.
Pyle, R. L., Barik, S. K., Christidis, L., Conix, S., Costello, M. J., van Dijk, P. P., Garnett, S. T.,
Hobern, D., Kirk, P. M., & Lien, A. M. (2021). Towards a global list of accepted species V. The
devil is in the detail. Organisms Diversity & Evolution, 1–19.
Rheindt F.E., Bouchard P., Pyle R.L., Welter-Schultes F.., Aescht E., Ahyong S.T., et al.
(2023). Tightening the requirements for species diagnoses would help integrate DNA-based
descriptions
in
taxonomic
practice.
PLoS
Biol
21(8):
e3002251.
https://doi.org/10.1371/journal.pbio.3002251
Riddle, B. R., Ladle, R. J., Lourie, S. A., & Whittaker, R. J. (2011). Basic Biogeography:
Estimating
Biodiversity
and
Mapping
Nature.
Conservation
Biogeography.
https://doi.org/10.1002/9781444390001.ch4
Sandall, E. L., Maureaud, A. A., Guralnick, R., McGeoch, M. A., Sica, Y. V., Rogan, M. S.,
Booher, D. B., Edwards, R., Franz, N., Ingenloff, K., Lucas, M., Marsh, C. J., McGowan, J.,
Pinkert, S., Ranipeta, A., Uetz, P., Wieczorek, J., & Jetz, W. (2023). A globally integrated
structure of taxonomy to support biodiversity science and conservation. Trends in Ecology &
Evolution. https://doi.org/10.1016/j.tree.2023.08.004
Sangster, G. (2009). Increasing numbers of bird species result from taxonomic progress, not
taxonomic inflation. Proceedings of the Royal Society B: Biological Sciences, 276(1670),
3185–3191.
53

Sibley, C. G., & Monroe, B. L. (1990). Distribution and Taxonomy of Birds of the World. Yale
University Press.
Simkins, A. T., Buchanan, G. M., Davies, R. G., & Donald, P. F. (2020). The implications for
conservation of a major taxonomic revision of the world’s birds. Animal Conservation, 23(4),
345–352.
Stankowski, S., & Ravinet, M. (2021). Quantifying the use of species concepts. Current
Biology, 31(9), R428–R429.
Stropp, J., Ladle, R. J., Emilio, T., Lessa, T., & Hortal, J. (2022). Taxonomic uncertainty and
the challenge of estimating global species richness. Journal of Biogeography, 49(9), 1654–1656.
https://doi.org/10.1111/jbi.14463
Thiele, K. R., Conix, S., Pyle, R. L., Barik, S. K., Christidis, L., Costello, M. J., van Dijk, P. P.,
Kirk, P., Lien, A., & Thomson, S. A. (2021). Towards a global list of accepted species I. Why
taxonomists sometimes disagree, and why this matters. Organisms Diversity & Evolution,
21(4), 615–622.
Thomson, S. A., Thiele, K., Conix, S., Christidis, L., Costello, M. J., Hobern, D., Nikolaeva, S.,
Pyle, R. L., van Dijk, P. P., & Weaver, H. (2021). Towards a global list of accepted species II.
Consequences of inadequate taxonomic list governance. Organisms Diversity & Evolution,
21(4), 623–630.
Troudet, J., Grandcolas, P., Blin, A., & Vignes-lebbe, R. (2017). Taxonomic bias in biodiversity
data and societal preferences. Scientific Reports, 7(9132), 1–14. https://doi.org/10.1038/s41598017-09084-6
Vaidya, G., Lepage, D., & Guralnick, R. (2018). The tempo and mode of the taxonomic
correction process: How taxonomists have corrected and recorrected North American bird
species over the last 127 years. PLoS ONE, 13(4), 1–19. https://doi.org/https://
doi.org/10.1371/journal.pone.0195736
Wheeler, Q. (2020). A taxonomic renaissance in three acts. Megataxa, 1(1), 4–8.
Whittaker, R. J., Ara, xfa, jo, M. B., Jepson, P., Ladle, R. J., James, E. M. W., Willis, K. J.,
Araújo, M. B., Jepson, P., Ladle, R. J., Watson, J. E. M., & Willis, K. J. (2005). Conservation
Biogeography:
Assessment
and
Prospect.
Diversity
and
Distributions.
https://doi.org/10.1111/j.1366-9516.2005.00143.x
Williams, P. H. (2022). Novel splitting/lumping index reflects the history of species concepts
applied to bumblebees (Insecta: Apidae). Zoological Journal of the Linnean Society, 196(2),
704–719. https://doi.org/https://doi.org/10.1093/zoolinnean/zlab123
Zachos, F. E. (2016). Species Concepts in Biology. Springer International Publishing.
https://doi.org/10.1007/978-3-319-44966-1
Zachos, F. E. (2018a). (New) Species concepts, species delimitation and the inherent limitations
of taxonomy. Journal of Genetics, 97(4), 811–815.
Zachos, F. E. (2018b). Mammals and meaningful taxonomic units: the debate about species
concepts and conservation. Mammal Review, 48(3), 153–159.
Zapata, F. A., & Ross Robertson, D. (2006). How many species of shore fishes are there in the
Tropical
Eastern
Pacific?
Journal
of
Biogeography,
34(1),
38–51.
https://doi.org/10.1111/j.1365-2699.2006.01586.x
54

Capítulo 2
Revista: Global ecology and biogeography
Status: Finalizado para submissão
Do biological and ecological variables influence nomenclatural
uncertainty in birds?
Thainá Lessaa*, Fernanda Alves-Martinsb,c, Javier Martinez Arribasb,c, Karoline
Azevedoa, Ana C.M. Malhadoa, Juliana Stroppd,e, Richard J. Ladlea
a

Institute of Biological and Health Sciences, Federal University of Alagoas, 57072-970

Maceió, Alagoas, Brazil.
b

CIBIO-InBIO, Research Centre in Biodiversity and Genetic Resources, University of

Porto, Campus de Vairão, 4485-661 Vairão, Portugal.
c

BIOPOLIS Program in Genomics, Biodiversity and Land Planning, CIBIO, Campus de

Vairão, 4485-661 Vairão, Portugal
d

Department of Biogeography and Global Change, Museo Nacional de Ciencias

Naturales (MNCN-CSIC), Madrid, Spain
e

Department of Biogeography, Trier University, Trier, Germany.

*Corresponding author: Thainá Lessa;
Email: thainalessaps@gmail.com

55

Abstract
Aim: Taxonomic nomenclature is the system by which scientists standardise and
communicate unambiguously about organisms. However, even well-studied taxa such
as birds there are taxonomic inconsistencies, due to divergences in the concept of
species. As a result, there are multiple lists of birds, with different estimates of species
number. Information on which characteristics of birds influence the degree of
taxonomic uncertainty is incipient. Here we create a metric to calculate nomenclatural
uncertainty exploring biases in bird species and Orders. Further, we analyse associations
between our nomenclatural uncertainty metric and biological and ecological variables of
bird species.
Location: Global.
Time period: Present.
Major taxa studied: Birds.
Methods: We used the International Ornithological Community (IOC) World Bird List
which compiles and compares among other global bird lists the scientific names of
species around the world. We created a nomenclatural uncertainty score for 11,140 bird
species, using the proportion of disagreement (and absence) of scientific names between
nine world bird lists. We assess drivers of variation in our nomenclatural uncertainty
metric from bird’ biological and ecological variables: body mass, range size, habitat
density, lifestyle, IUCN status, and evolutionary distinctiveness.
Results: More than 50% of global bird species presented some degree of nomenclatural
uncertainty, and for bird Orders the percentage was greater than 80%. Birds with
smaller body size, smaller geographic range, classified by the IUCN as threatened and
deficient data, and with low evolutionary distinctiveness were the most affected by
nomenclatural uncertainty.
Main conclusions: Biological-ecological variables such as body mass and range size are
strictly related to the accessibility and convenience of taxonomists to collect and
describe species. Larger, more widely distributed, and more common species are easier
to observe and collect. Species with characteristics contradictory to these should be
priorities for assessments of taxonomic stability.
Key-words: taxonomy, nomenclature, linnean shortfall, conservation, ornithology

56

Introduction
Taxonomy is the first step of most systematic biology and natural history studies, and
underpins our understanding of biodiversity (Agnarsson & Kuntner, 2007). As the
fundamental and most frequently used taxonomic unit, ‘species’ are universally used in
studies of biodiversity, ecology, evolution, biogeography and conservation (Agapow et
al., 2004; Barrowclough et al., 2016; De Queiroz, 2007; Isaac et al., 2004; Ladle &
Whittaker, 2011). However, in a dynamic and complex nature world achieving
comprehensive knowledge about any aspect of species remains largely impracticable
(Ladle & Hortal, 2013). Disagreements over the taxonomic classification and identity of
species can have significant impacts on fundamental characteristics such as the number
of known and unknown species. The gap in our knowledge of species identities is
referred to as Linnean shortfall (Brown & Lomolino, 1998) and encompasses both
species that have not yet been discovered in nature, and species that have been collected
but not formally named (Hortal et al., 2015). The latter case includes specimens in
museum collections that have never been adequately assessed (Bebber et al., 2010;
Hortal et al., 2015), but also species currently included within another species – often
recognized as subspecies or another taxonomic category – that will be upgraded to full
species in the future through re-evaluation using new approaches (Parsons et al., 2022).
This is further complex by the fact that species are hypotheses (Pante et al., 2014), and
proposed changes in taxonomic status are not necessarily accepted or adopted by the
scientific community.
It would be recommended that organisms collected be unambiguously described
and named prior to further analysis (Riddle et al., 2011). However, the scientific
consensus about how to identify and delimit species has changed dramatically over the
last decades. The current taxonomic landscape is characterised by several distinct (and
viable) concepts to define what species are (Kitchener et al., 2022; Zachos, 2016, 2018),
and a large number of methodologies and criteria that are used to delimit it (Carstens et
al., 2013; Hillis et al., 2021; Rannala, 2015). Modern approaches stress the importance
of integrative taxonomy, combining genetic data with phenotypic, behavioral, and
ecological traits to identify significant discontinuities and reproductively isolated
populations that merit species status (Cicero et al., 2021). Such an approach requires
high scientific capacity and technology, and extensive geographic sampling that
substantially captures variations in population structure including putative contact

57

zones. In summary, there remains a large amount of both conceptual and practical
uncertainty around species data that increases with spatial and temporal scale.
Even in well-studied taxa such as birds there is a constant state of flux with
knock on impacts for estimates of total species richness (known + unknown species)
(Jetz et al., 2012; Neate-Clegg et al., 2021). Taxonomic inconsistency is reflected by the
multiple bird lists worldwide, based on divergent taxonomic criteria (McClure et al.,
2020; Neate-Clegg et al., 2021). These multiple bird lists attend to distinct purposes,
have a singular importance for science and conservation, and are periodically reviewed
by groups of experts based on a series of species definitions (Garnett & Christidis,
2007; Hey et al., 2003; Thomson et al., 2018), a methodology that can lead to
discordant estimates of species richness and taxonomic uncertainty. A recent study
found that 68% of four bird lists exhibit some degree of discordance, whether through
taxonomic omissions, the use of different scientific name (full name, epithet, and/or
genus), or treatment at the taxonomic level (species and subspecies) (McClure et al.,
2020). Dealing with this challenge is crucial and requires commitments to the original
conceptualization of species. The need for greater standardisation of data and
methodologies to address how species are identified and catalogued is well recognized,
and there are recent proposals for how this might eventually be dealt with (e.g., Cicero
et al., 2021; Garnett et al., 2020; Orr et al., 2022). Until this happens, it is important that
scientists better understand the causes and consequences of taxonomic uncertainty so
that it might be incorporated into biogeographic and conservation models (Tessarolo et
al., 2021).
One approach to this is to investigate the biocultural characteristics of
taxonomically disputed/uncertain species. Here, it is important to distinguish between
uncertainty about the identity of an observation or specimen (e.g. species A is
mistakenly identified and recorded as species B) and uncertainty about the validity of a
taxonomic identification. The latter can occur when the taxonomic status of a species is
disputed (e.g. some scientists consider it a sub-species, others consider it a valid
species) or when there are multiple unresolved taxonomies, leading to synonyms.
Studying the drivers of taxonomic uncertainty thus faces the considerable challenge of
distinguishing between and quantifying degrees of certainty and consensus for species
level biodiversity data. All things being equal the probability of mistaken identity
increases with decreasing levels of taxonomic/natural history expertise, and level of
experience of the observer has been used as a proxy of taxonomic certainty in
58

biogeographic studies (Tessarolo et al., 2021). Lower taxonomic categories (e.g. genera
and species) are more problematic and have typically been evaluated through measures
of nomenclatural stability. For example, degree of consensus among global species
checklists, each one reflecting the taxonomic judgement of a subset of the global
scientific community (Hobern et al., 2021).
Recognizing and quantifying gaps in knowledge about the species’ identity is
imperative, as distorted data can mislead our understanding of biodiversity descriptions
and their ecological and evolutionary processes (Cardoso et al., 2011; Hortal et al.,
2015; Stropp et al., 2022). A striking example is highlighted in global assessments of
bird species numbers, which range from 10,000 to 18,000 depending on the application
of species concept (Barrowclough et al., 2016; Mayr, 1946; Neate-Clegg et al., 2021).
This disparity shows the importance of addressing taxonomic uncertainties to enhance
the accuracy and reliability of biodiversity estimates. The first step is to assess these
official lists of species in an attempt to make them a unique global reference (Lien et al.,
2021) evaluate whether the intrinsic characteristics of birds influence uncertainties.
Here, we compare scientific name disagreement between world bird lists using the
International Ornithological Community (IOC) World Bird List, version 13.1 as a
reference list. We create a nomenclatural uncertainty score using the proportion of
disagreement (and absence) of scientific names between the world bird lists, and we
explore biases in bird species and Orders. Finally, we analyse statistical associations
between biological and ecological variables of bird species and our nomenclatural
uncertainty score.
Methods
World bird lists
We downloaded the International Ornithological Community (IOC) World Bird List
version 13.1 (Gill et al., 2023) (hereafter IOC 2023.1) in November 2023. The IOC
2023.1 is an open access database that compiles the scientific names of birds species
and subspecies globally (living and extinct) based on evolutionary classification (Gill et
al., 2024), and compares it with two previous versions (IOC 2022.1 and IOC 2022.2)
and other eight bird lists: eBird/Clements (2022), HBW and BirdLife International
(2022), Jimmy Gaudin (2021-2022), John H. Boyd (2019-2021), Howard and Moore
(2014), Sibley and Monroe (1993); del Hoyo et al. (1992-2013), Peters et al. (19311986).
59

The IOC 2023.1 contains 30,971 bird species and subspecies, with 1,031 bird
species mentioned in other lists not recognized by IOC 2023.1. As a taxonomic
delimitation, we filtered only ‘species’ classification, resulting in 11,140 bird species.
We made two types of comparisons: a) between the numbers of species among 11 world
bird lists: we conducted pairwise comparisons between the bird lists to assess the
percentage of agreement in the number of bird species (i.e. those species that had the
same scientific names) across these 11 world bird lists, using IOC 2023.1 as reference
list. In disagreement cases, we evaluated where it occurred in the scientific name, for
example, discordances in the generic name (genus), specific name (epithet) or complete
name (Supplementary material 1); b) between the agreement/disagreement/absence of
scientific names filled in nine world bird lists; this last database was used for
nomenclature uncertainty calculations (see next section).
Taxonomic uncertainty score
We compute scientific name disagreement (and absence) of birds among world bird lists
as a proxy for nomenclatural uncertainty. As our preliminary results showed that IOC
2022.1 and IOC 2022.2 are extremely similar to IOC 2023.1 (>99%), we removed them
from the calculation to avoid underestimation. We performed nomenclatural uncertainty
calculations in Microsoft Excel and did it for all 11,140 bird species listed in IOC
2023.1. We made the calculation by comparing whether the scientific name listed in the
IOC 2023.1 was the same, different or if was not mentioned in the other eight lists
(mentioned in the section above) (Table 1). For this, we first counted the number of
filled cells with a scientific name in the range of the nine lists evaluated (each cell was a
list: IOC 2023.1 and eight bird lists), using the scientific name listed in IOC 2023.1 as a
counting criteria. The outcome was zero when the cell (list) was not filled with a
scientific name. The results of these counts were summed and then divided by the
difference of the total number of lists evaluated squared and the multiplication of the
count of the number of lists evaluated and the number of lists not filled with a scientific
name. Finally, the final values were subtracted by one to obtain nomenclatural
uncertainty. We transformed the final values into percentages for a more intuitive
representation (see Equation 1):

60

Where

is the sum of the number of lists with a scientific name, obtained by

counting from each one of the nine bird lists with a present (agreed or disagreed) or
absent scientific names based on the IOC 2023.1 list; n is the number of lists evaluated
(in our case this was nine); n0 is the number of lists not filled with a scientific name.
This equation presents a comprehensive methodology for evaluating nomenclatural
uncertainty, encompassing considerations of both unfilled list occurrences and the
variety of scientific names within each list. Once we recognized that there are two
contexts with different outcomes: scientific names absent in the lists and scientific
names in disagreement based on a reference list. The final expression yields a
percentage measure of nomenclatural uncertainty, facilitating interpretation and
comparison across different taxonomic scenarios.
Table 1: Examples of nomenclatural uncertainty among scientific names on world bird lists.
eBird/
Clements
(2022)

HBW
and
BirdLife
Intern.
(2022)

Howard
and
Moore
(2014)

del Hoyo
et al.
(19922013)

Struthio
camelus

Struthio
camelus

Struthio
camelus

Struthio
camelus

Struthio
camelus

Struthio
camelus

Struthio
camelus

Struthio
camelus

Struthio
camelus

Pardirallus
nigricans

Pardirallus
nigricans

Pardirallus
nigricans

Pardirallus
nigricans

Pardirallus
nigricans

Ortygonax
nigricans

Pardirallus
nigricans

Pardirallus
nigricans

Pardirallus
nigricans

Ardea
alba

Ardea
alba

Ardea
alba

Ardea
alba

Egretta
alba

Ardea
alba

Casmerodius
albus

Casmerodius
albus

Casmerodius
albus

Nesotriccus
tumbezanus

Nesotriccus
tumbezana

Phaeomyias
tumbezana

Phyllomyias
tumbezanus

Nesotriccus
tumbezanus

IOC
2023.1

Peters
et al .
(19311986)

John H.
(20192021)

Jimmy
Gaudin
(20212022)

Sibley and
Monroe
(1993)

To illustrate the outputs, if the species has 100% of agreement between the nine
bird lists, consequently has 0% of nomenclatural uncertainty. Therefore, nomenclatural
uncertainty was considered null when there was complete agreement (uncertainty = 0%)
and varying when there was disagreement or absence in scientific name between the
lists (uncertainty ≠ 0%). For example, Struthio camelus had the same scientific name in
all nine lists (including IOC 2023.1), so its nomenclatural uncertainty was 0%; In one of
nine list, Pardirallus nigricans had a discordant scientific name, so this species had 20%
of taxonomic uncertainty; Ardea alba had a discordant scientific name in four lists then
there was 57% uncertainty associated with its scientific name; and Nesotriccus
tumbezanus had its scientific name equal in one list, absent scientific name in four lists
61

and discordant scientific name in three lists, resulting in 84% of nomenclatural
uncertainty (Table 1; see detailed list in Supplementary material 2). In addition to
assessing nomenclatural uncertainty across bird species, we also examined the
distribution of nomenclatural uncertainty across bird Orders.
Biological and ecological variables of birds
To understand what drives the variation in nomenclatural uncertainty, we collected
biological and ecological variables (Table 2) based on IOC 2023.1 bird list. Firstly, we
cross-referenced the IOC 2023.1 list with the AVONET list - a database published in
2022 that promotes detailed curation on global bird’s traits, ecology and biogeography
(Tobias et al., 2022). AVONET database presents separate information from three bird
lists: BirdLife International, eBird and BirdTree. We used data from BirdLife
International as it contains the largest number of species (n=10,999). We selected
information on body mass, range size, habitat density, lifestyle and IUCN conservation
status. We group IUCN’ status into three categories: i) Not threatened (NTR):
encompasses both Least Concern (LC) and Near Threatened (NT); ii) Threatened (TR):
comprises Vulnerable (VU), Endangered (EN) and Critically Endangered (CR); and iii)
Data Deficient (DD). Nomenclatural uncertainty calculations were carried out for
Extinct (EX) or Extinct in the Wild (EW) bird species, but they were not included in the
analysis model. The variable habitat density has three categories: dense, semi-open and
open habitats. The lifestyle variable has five categories: Aerial, Aquatic, Insessorial,
Terrestrial and Generalist. As a result we obtained data for 9,994 bird species listed in
IOC 2023.1. Secondly, we collected data on the evolutionary distinctiveness (ED) of
bird species from EDGE of Existence, a Zoological Society of London's conservation a
program that supports conservation and research for species both evolutionarily distinct
and globally threatened (Faith, 2019). This database provides data on 10,954 species,
the ED Scores were used. Combining data from IOC 2023.1, AVONET and EDGE of
Existence, was returned complete information on 9,864 bird species.
We model these biological and ecological variables as a function of
nomenclatural uncertainty. Given that our response variable (nomenclatural uncertainty)
represents a proportion, we conducted a beta regression using the 'betareg' package
(Cribari-Neto & Zeileis, 2010). Prior to the analysis, we added 10e-06 to nomenclatural
uncertainty scores to prevent zeros, as the beta distribution requires values greater than
zero (for a similar approach see Lessa et al., 2024).
62

Table 2: Biological and ecological variables used to explain nomenclatural uncertainty among
bird species. The table provides a brief assumption of why variables were included into the
model and the source of data collected.
Variables

Justification

Format

Source

Body Mass

Small species are more difficult to collect (Hey et
al., 2003), consequently present greater
nomenclatural uncertainty.

Continuous

Avonet

Range Size

Species with restricted distribution area are more
challenging and expensive to research (Moura &
Jetz, 2021), so have greater taxonomic uncertainty.

Continuous

Avonet

Categoric

Avonet

Lifestyle

Generalist species are more common in urban
environments and easier to observe (Callaghan et al.,
Categoric
2019), therefore they present lower taxonomic
uncertainty.

Avonet

IUCN

Threatened species attract less research interests
(Jarić et al., 2019), therefore have greater
nomenclatural uncertainty

Categoric

Avonet

Evolutionary
distinctiveness

Species with greater evolutionary distinctiveness
are more studied (dos Santos et al., 2020), thus have
lower nomenclatural uncertainty.

Continuous

EDGE

Species from dense habitats are harder to study
Habitat Density (Neate-Clegg et al., 2021), thus have greater
nomenclatural uncertainty.

Results
We analysed the number of species described in world bird lists and the degree of
agreement between their scientific names in comparison to the 11,140 bird species from
the reference list, IOC 2023.1. The lists that showed the highest percentage between the
numbers of bird species with agreed scientific names by IOC 2023.1 were its previous
versions IOC 2022.2 and IOC 2022.1, both showing 99% of agreement in scientific
names (Figure 1). Following, the most recent lists with the largest number of species
were also those that achieved the highest percentage of agreement in scientific names:
eBird/Clements (2022) (95%) and HBW and BirdLife International (2022) (93%). On
the other hand, the lists with the greatest disagreements in scientific names were the
oldest with the lowest number of species: Peters (1931-1986) with more than 50% of
disagreements in relation to IOC 2023.1, and Sibley & Monroe (1993) with 65%
(Figure 1).
The majority of sources of disagreement in the scientific names in all lists were
divergence in the generic name and full name. However there was also discordance in
63

the specific name, especially in the oldest lists: Peters (1931-1986), Sibley & Monroe
(1993) and del Hoyo (1992-2013). For example, from 46% of bird species with
scientific names in disagreement in Peters (1931-1986) list, 1,812 species had a generic
name disagreed, 424 species had a specific name disagreed and 388 species had a
scientific name completely disagreed. In Sibley & Monroe (1993) list 35% of bird
species had disagreement in scientific names, being 1,718 species with disagreement in
generic name, 323 species disagreement in specific name, and 326 species disagreement
in full scientific name. In previous versions of IOC 2023.1, IOC 2022.1 had six species
with a completely different scientific name, 33 species that had a generic name
disagreed and three species with a specific name disagreed; and IOC 2022.2 also had six
species with a completely scientific name disagreed and 17 species had their generic
name disagreed (Figure 1).

Figure 1: Bar chart representing the nomenclatural agreement of extant world bird lists in
comparison with a reference list (IOC 2023.1). Dark blue bars represent the number of
completely discordant species (genus and species different), medium blue represents change in
genus (medium blue) and light blue represents a change in species name. The number next to
the bird silhouette refers to the total number of bird species in each list. The percentages in the
graph have been rounded to the nearest whole number. For full values, see Supplementary
material 1.

From our nomenclatural uncertainty calculation, 36 bird Orders were evaluated
(Figure 2). Only six bird Orders did not have nomenclatural uncertainty, i.e. all bird
species had scientific names that agreed among all nine lists used in the metric. This
bird Orders were also those had the smallest number of species: Cariamiformes (n= 2),
Cathartiformes

(n=

7),

Eurypygiformes

(n=

2),

Leptosomiformes

(n=

1),

Opisthocomiformes (n= 1) and Phaethontiformes (n= 3). The Orders that had the
64

highest mean of number of species with nomenclatural uncertainty were: Pterocliformes
(n= 29) with 81% of the birds in this Order presenting nomenclatural uncertainty that
ranged between 35% to 37% (mean of 36%); Otidiformes (n= 47) with 80% of the birds
in this Order had nomenclatural uncertainty ranging between 20% to 69% (mean of
37%); Suliformes (n= 109) with 78% of the birds in this Order showed nomenclatural
uncertainty varying between 20% to 78% (mean of 47%); Musophagiformes (n= 40)
about 74% of the birds in this Order presented nomenclatural uncertainty that ranged
between 11% to 64% (mean of 36%); Strigiformes (n= 443) almost 72% of the birds in
this Order exhibited nomenclatural uncertainty ranging between 11% to 89% (mean of
37%); and Bucerotiformes (n= 126) with 70% of the birds in this Order presenting
nomenclatural uncertainty that ranged between 11% to 67% (mean of 34%).

Figure 2: Column graph representing the percentage of the number of species with
nomenclatural uncertainty in bird Orders. The table below is the total number of species (green)
and number of species with nomenclatural uncertainty (red) per Order. Boxplot (orange) shows
the distribution of nomenclatural uncertainty in the Orders. Accipitriformes (ACC),
Anseriformes (ANS), Bucerotiformes (BUC), Caprimulgiformes (CAP), Cariamiformes (CAR),
Cathartiformes (CAT), Charadriiformes (CHA), Ciconiiformes (CIC), Coliiformes (COL),
Columbiformes (CLB), Coraciiformes (COR), Cuculiformes (CUC), Eurypygiformes (EUR),
Falconiformes (FAL), Galliformes (GAL), Gaviiformes (GAV), Gruiformes (GRU),
Leptosomiformes
(LEP),
Mesitornithiformes
(MES),
Musophagiformes
(MUS),
Opisthocomiformes (OPI), Otidiformes (OTI), Passeriformes (PAS), Pelecaniformes (PEL),
Phaethontiformes (PHA), Phoenicopteriformes (PHO), Piciformes (PIC), Podicipediformes
(POD), Procellariiformes (PRO), Psittaciformes (PSI), Pterocliformes (PTE), Sphenisciformes
(SPH), Strigiformes (STR), Struthioniformes (STT), Suliformes (SUL), Trogoniformes (TRO).

Further about our nomenclatural uncertainty calculation for the 11,140 bird
species evaluated, over half species (56%) had some degree of nomenclatural
uncertainty, varying from 11% to 89% (see Supplementary material 2). More than 16%
65

(n=1,812) of bird species had a nomenclatural uncertainty score greater than 50%. The
birds species with the highest nomenclatural uncertainty score (89%) were: Centropus
burchellii (Cuculiformes), Cinnyris whytei (Passeriformes) and Ramosomyia wagneri
(Caprimulgiformes) which were only present on the IOC 2023.1 list and on another list
with a scientific name in disagreement, and Loriculus bonapartei (Psittaciformes), Otus
bikegila (Strigiformes), Phylloscopus floresianus (Passeriformes), Setophaga aestiva
(Passeriformes) and Zosterops paruhbesar (Passeriformes) which were only present on
the IOC 2023.1 list. Other bird species with a high nomenclatural uncertainty score
(>81%)

were:

Nesotriccus

(Columbiformes),

Arborophila

tumbezanus
diversa

(Passeriformes),
(Galliformes),

Ptilinopus

Melanerpes

gularis

santacruzi

(Piciformes), Psittacara brevipes (Psittaciformes) and Ramosomyia viridifrons
(Caprimulgiformes).
The results of our model to assess the relationship between nomenclatural
uncertainty score and biological and ecological variables of birds (Table 3) revealed that
the body mass was associated with nomenclatural uncertainty, with a trend to the
nomenclatural uncertainty being lower in larger bird species. The species distribution
area was also associated with nomenclatural uncertainty score, which nomenclatural
uncertainty was lower in bird species with a wider range size. Evolutionary
distinctiveness, i.e. species from an unusual group with a remarkable evolutionary
history, likewise showed an association with nomenclatural uncertainty, with
uncertainty being lower in more distinctive bird species. Finally, IUCN conservation
status was associated with nomenclatural uncertainty, and was greater in bird species
categorised as Threatened (VU, EN and CR). The habitat density and lifestyle variables
were not significant.
Table 3: Significant results from the Beta regression model analysing the association between
nomenclatural uncertainty and biological and ecological variables of bird species. Full results
with non-significant associations are available in Supplementary material 3.
Variables
Body Mass
Range Size
Evol.Distinctiveness
IUCN Threatened

Coefficient
estimate
-1.123e-05
-5.549e-03
-9.964e-03
1.915e-01

z value

Pr(>|z|)

-2.031
-3.779
-4.434
5.485

0.042208 *
0.000158 ***
9.23e-06 ***
4.13e-08 ***

66

Discussion
Birds are one of the most studied classes of vertebrate fauna worldwide and recognized
to show a taxonomic stability (Barrowclough et al., 2016; Neate-Clegg et al., 2021).
However, there are still major gaps and deficiencies regarding the classification and
nomenclature of this group. Our results revealed that there are great divergences in the
number of species in each bird list evaluated, with the International Ornithological
Community (IOC) lists presenting the highest species count. The IOC carries out
biannual updates of taxonomic reviews, and the newest update published in 2024
already counts an increase in the number of bird species (55 new species) (Gill et al.,
2024). On the other hand, older bird checklists such as Peters (1931-1986), Sibley &
Monroe (1993) and del Hoyo (1992-2013) set fewer species numbers, which is not
surprising due to technological limitations to describe and name species formerly.
Nonetheless, the existence of these checklists demonstrates the importance of
pioneering studies for the improvement of more complete studies/lists, such as the
creation of the eBird/Clements and the Howard and Moore lists that were based on
Peters list (1931-1986) (Lepage & Warnier, 2014).
The checklist from the citizen science platform eBird (eBird/Clements (2022))
and the list used by the International Union for Conservation of Nature (IUCN) (HBW
and BirdLife International (2022)) also exhibit a high number of species and similarity
of scientific names in comparison to the IOC 2023.1 list, with 95% and 93% of
agreement, respectively. These discrepancies, to a large or small degree, are most likely
related to the species concept applied by each list. For example, the IOC list uses the
evolutionary species concept, while the eBird and, HBW and BirdLife International lists
applies the biological species concept (McClure et al., 2020). There is already evidence
that when different species concepts are applied occur changes in estimates of species
richness, and even in patterns of endemism. For sub-Saharan African birds there was a
33% increase in the number of species using the phylogenetic species concept (n=
2,098) in relation to the biological species concept (n= 1,572) (Dillon & Fjeldså, 2005).
The same standard was observed for birds from Mexico, which after literature and
museum

specimens

review,

135

biological

species

were

split

into

323

phylogenetic/evolutionary species, increasing 125% in the number of species endemic
to Mexico (Navarro-Sigüenza & Peterson, 2004; Peterson & Navarro-Sigüenza, 1999).
In our study, more than half (56%) of global bird species showed some degree of
nomenclatural uncertainty associated with differences in the scientific name among nine
67

global bird lists. The scenario becomes more critical in higher taxonomic levels, with
more than 80% of bird Orders holding nomenclatural uncertainty. The factors of
disagreement in scientific names were especially linked to processes of splitting,
lumping and taxonomic swap. These processes, especially splitters, can result in
‘taxonomic inflations and deflations’, i.e., increase and decrease in the number of
species by taxonomic changes and delimitations (Isaac et al., 2004). Debates are intense
about whether taxonomic inflation/deflation is real or it is simply a reflection of
taxonomic progress (Garnett & Christidis, 2017; Harris & Froufe, 2005; Sangster, 2009;
Wheeler, 2020). Perhaps this widespread concern among ornithologists could be better
addressed by understanding and appropriately applying integrative taxonomy – that is to
include multiple lines of evidence that are useful when inferring reproductive isolation
to determine species limits (Mandiwana-Neudani et al., 2021; Winker & Rasmussen,
2021). In any case, taxonomic changes need to be recognized and incorporated into
biodiversity metrics (Stropp et al., 2022)).
The most marked divergences between the scientific names across bird lists
evaluated here were in the generic name (genus), full scientific name and specific name
(epithet), although the latter was less prominent. This trend should be also a reflection
of taxonomic progress that began to include molecular data and phylogenetic
relationships in species descriptions, leading to an increase in the number of genera, as
observed in ferns (Christenhusz & Chase, 2018). Another explanation is that
taxonomists may prefer to split genera into smaller groups to facilitate and detail
ecological and evolutionary knowledge of species (Gasper, 2016). Although
international codes of nomenclature stipulate rules, there are no objective criteria of
what constitutes a category (e.g., genera or family) in biological classification.
Taxonomists are free to choose scientific names, based on available data, and propose it
for acceptance (or not) to the scientific community (Rheindt et al., 2023). Perhaps
taxonomy will never reach a definitive consensus due to the dynamics and constant
evolution of nature. But rather also for human characteristics, since the designation of
what a species is, is part of human judgement and culture. Even if there was a single
species concept and scientists worked with the same dataset, there would be
disagreements (Conix et al., 2023).
Some trends explain why many species are awaiting to be discovered while
others have already undergone several taxonomic reviews (Alroy, 2003; May, 1988;
Moura & Jetz, 2021). As revealed in our model, the bird species most affected by
68

nomenclatural uncertainty were those that had smaller body size, smaller geographic
distribution (i.e. restricted or endemic to a region), species classified by the IUCN as
threatened and deficient data, and species with low evolutionary distinctiveness. For
example, Otus bikegila (Strigiformes) is an owl critically endangered and endemic to
Principe Island in São Tomé and Príncipe, and Zosterops paruhbesar (Passeriformes) is
a small passerine listed as endangered and restricted to Wangi-wangi Island, Indonesia,
both species were recently described after careful review using a huge of evidences
(Irham et al., 2023; Melo et al., 2022).
Furthermore, several species with the highest nomenclatural uncertainty scores
were considered subspecies among the lists, such as Phylloscopus floresianus
(Passeriformes) which is restricted to Timor-Leste and is considered a subspecies
(Phylloscopus presbytes floresianus and Phylloscopus presbytes floris) in other lists (see
Supplementary material 2). The debate surrounding the utility of subspecies has
persisted for decades, with the controversial principle that they are critical stages in
evolution, forming part of a continuum from limited differentiation among populations
to reproductive isolation (De Queiroz, 2007). However, many subspecies often require
re-examination to assess their validity, as they were described decades ago based on few
specimens and characters, with nomenclature applied to populations that are poorly
diagnosable (Cicero et al., 2021; Winker & Haig, 2010), which could be resulting in
these higher rates of nomenclatural uncertainty.
Two biological and ecological variables are strictly related to the accessibility
and convenience of taxonomists in collecting and describing species, such as body size
and geographic range (Clark & May, 2002; Ladle & Whittaker, 2011; Moura & Jetz,
2021). For example, for Australian birds, more scientific publications are found for
those with larger body size, wide geographic distribution and abundance (Yarwood et
al., 2019). Furthermore, a similar study evaluating levels of agreement between four
bird lists and ecological variables demonstrated that agreement was greater in larger
species (Neate-Clegg et al., 2021). The premise behind this is that larger species are
easier to observe and collect, but are also correlated with lower diversity, consequently
presenting lower taxonomic uncertainties (Alroy, 2003; dos Santos et al., 2020; Hey et
al., 2003). Similarly, species with wide geographic range are also more likely to be
collected and studied, as their distribution often overlaps with the distribution of
research groups/taxonomists (dos Santos et al., 2020; Meyer et al., 2015; Yarwood et
al., 2019), and therefore, are less likely to have nomenclatural uncertainty.
69

A similar bias was observed for bird Orders, although we did not evaluate it
statistically. Birds from Orders that presented the greatest number of species with
nomenclatural uncertainty share some biological and ecological characteristics, which
may be hindering field collection and taxonomic investigation. For example, species
belonging to Pterocliformes (sandgrouse) are small in size and range, and have cryptic
plumage. Otidiformes (bustards) species also have cryptic plumage, solitary behaviour,
and live in savannas and/or arid landscapes. And Strigiformes species, such as owls,
although they have a wide geographic range, have solitary and nocturnal behaviour.
Our nomenclatural uncertainty metric has a number of unavoidable limitations.
Firstly, it is important to acknowledge that taxonomy is a dynamic, continually evolving
science and that our analysis is only a ‘snapshot’ of the current level of nomenclatural
discordance. As the taxonomic system evolves further discordance will be generated
and, at the same time, older species lists will fall into obsolescence. Alternatively, the
current taxonomic and nomenclatural system may be revised and a single, authoritative
global species list may emerge (Lien et al., 2021; Pyle et al., 2021) and become
universally adopted. Until that time, duplicated efforts and confusion in the scientific
use of taxonomic information will continue to breed nomenclatural uncertainty along
with its negative consequences for ecologically and conservation research (Thomson et
al., 2021). It is also important to note that the status of some taxonomic units will
always be contentious since, even when faced with the same data, experts may disagree
about the taxonomic status of groups in the process of speciation (Conix et al., 2023).
Previous studies have already carried out investigations into taxonomic quality,
such as the binary metric proposed by Neate-Clegg et al., (2021) that assesses
agreement between taxonomic authorities for global birds; and the indicator of
identifiers’ expertise (taxonomists, experts and amateurs) on the quality of taxonomic
identifications for Scarabaeidae dung beetle species (Tessarolo et al., 2021). However,
the

aforementioned

metrics

are

limited

by

not

detailing

the

degree

of

taxonomic/nomenclatural quality or/and needs to prior know regarding the types of
identifiers in the database, which may be impractical in global assessments. Our
nomenclatural uncertainty metric has the advantage of evaluating the degree of
nomenclatural uncertainty, in a broader and finer resolution, making bird species and
Order comparable, and using the available taxonomic data. In addition, it was possible
to assess the motivations behind the uncertainty, whether divergences in scientific
names and the biological and ecological characteristics of the species. Nevertheless, the
70

creation of our metric was only allowed due to the hard work carried out by researchers
and institutions, which compiled and made taxonomic information accessible.
Despite the immeasurable relevance of taxonomy as a basic science for
biodiversity studies, there are vast deficiencies that have led to a global crisis. On the
one hand, scientists are increasingly striving against the clock to collect and describe
species before they become threatened due to current environmental impacts worldwide
(Stropp et al., 2020). On the other hand, taxonomy is no longer seen with the 'glamour'
it once was, with there being less interest among young scientists, as well as there being
a shortage of taxonomists for many taxa (Engel et al., 2021). Especially due to the
scarce investments in the area and the bureaucracy involved from the description to the
publication of a new species (Britz et al., 2020). The requirement for additional
information beyond morphology increases the impediments to publishing articles on
review and descriptions of new species. As we saw here, approximately 12% of bird
species (not counting the subspecies) had no available data about biology and ecology.
Therefore, science will only fill the remaining empty boxes of knowledge if the Linnean
shortfalls are first filled (Hortal et al., 2015). Only when we join efforts and financing
actions that protect both ‘species’ in danger of extinction, will we move towards true
taxonomic progress (Wägele et al., 2011).
References
Agapow, P., Bininda‐Emonds, O. R. P., Crandall, K. A., Gittleman, J. L., Mace, G. M.,
Marshall, J. C., & Purvis, A. (2004). The Impact of Species Concept on Biodiversity Studies.
The Quarterly Review of Biology, 79(2), 161–179. https://doi.org/10.1086/383542
Agnarsson, I., & Kuntner, M. (2007). Taxonomy in a Changing World: Seeking Solutions for a
Science
in
Crisis.
Systematic
Biology,
56(3),
531–539.
https://doi.org/10.1080/10635150701424546
Alroy, J. (2003). Taxonomic inflation and body mass distributions in north american fossil
mammals. Journal of Mammalogy, 84(2), 431–443. https://doi.org/10.1644/15451542(2003)084<0431:TIABMD>2.0.CO;2
Barrowclough, G. F., Cracraft, J., Klicka, J., & Zink, R. M. (2016). How many kinds of birds
are there and why does it matter? PLoS One, 11(11), e0166307.
Bebber, D. P., Carine, M. A., Wood, J. R. I., Wortley, A. H., Harris, D. J., Prance, G. T.,
Davidse, G., Paige, J., Pennington, T. D., Robson, N. K. B., & Scotland, R. W. (2010). Herbaria
are a major frontier for species discovery. Pnas, 107(51), 22169–22171.
https://doi.org/10.1073/pnas.1011841108
Britz, R., Hundsdorfer, A., & Fritz, U. (2020). Funding, training, permits—the three big
challenges of taxonomy. Megataxa, 1(1). https://doi.org/10.11646/megataxa.1.1.10
71

Brown, J. H., & Lomolino, M. V. (1998). Biogeography.
Callaghan, C. T., Major, R. E., Wilshire, J. H., Martin, J. M., Kingsford, R. T., & Cornwell, W.
K. (2019). Generalists are the most urban‐tolerant of birds: a phylogenetically controlled
analysis of ecological and life history traits using a novel continuous measure of bird responses
to urbanization. Oikos, 128(6), 845–858. https://doi.org/10.1111/oik.06158
Cardoso, P., Erwin, T. L., Borges, P. A. V., & New, T. R. (2011). The seven impediments in
invertebrate conservation and how to overcome them. Biological Conservation, 144(11), 2647–
2655. https://doi.org/10.1016/j.biocon.2011.07.024
Carstens, B. C., Pelletier, T. A., Reid, N. M., & Satler, J. D. (2013). How to fail at species
delimitation. Molecular Ecology, 22(17), 4369–4383. https://doi.org/10.1111/mec.12413
Christenhusz, M. J. M., & Chase, M. W. (2018). PPG recognises too many fern genera.
TAXON, 67(3), 481–487. https://doi.org/10.12705/673.2
Cicero, C., Mason, N. A., Jiménez, R. A., Wait, D. R., Wang-Claypool, C. Y., & Bowie, R. C.
K. (2021). Integrative taxonomy and geographic sampling underlie successful species
delimitation. Ornithology, 138(2). https://doi.org/10.1093/ornithology/ukab009
Clark, J. A., & May, R. M. (2002). Taxonomic Bias in Conservation Research. Science,
297(5579), 191–192. https://doi.org/10.1126/science.297.5579.191b
Conix, S., Cuypers, V., Zachos, F. E., & De Block, A. (2023). Taxonomic disagreement about
ranks in gray-area taxa: A vignette study. BioScience, 73(10), 728–737.
https://doi.org/10.1093/biosci/biad081
Cribari-Neto, F., & Zeileis, A. (2010). Beta Regression in R. Journal of Statistical Software,
34(2). https://doi.org/10.18637/jss.v034.i02
De Queiroz, K. (2007). Species Concepts and Species Delimitation. Systematic Biology, 56(6),
879–886. https://doi.org/10.1080/10635150701701083
Dillon, S., & Fjeldså, J. (2005). The implications of different species concepts for describing
biodiversity patterns and assessing conservation needs for African birds. Ecography, 28(5),
682–692. https://doi.org/10.1111/j.2005.0906-7590.04344.x
dos Santos, J. W., Correia, R. A., Malhado, A. C. M., Campos‐Silva, J. V., Teles, D., Jepson, P.,
& Ladle, R. J. (2020). Drivers of taxonomic bias in conservation research: a global analysis of
terrestrial mammals. Animal Conservation, 23(6), 679–688. https://doi.org/10.1111/acv.12586
Engel, M. S., Ceríaco, L. M. P., Daniel, G. M., Dellapé, P. M., Löbl, I., Marinov, M., Reis, R.
E., Young, M. T., Dubois, A., Agarwal, I., Lehmann A., P., Alvarado, M., Alvarez, N.,
Andreone, F., Araujo-Vieira, K., Ascher, J. S., Baêta, D., Baldo, D., Bandeira, S. A., …
Zacharie, C. K. (2021). The taxonomic impediment: a shortage of taxonomists, not the lack of
technical approaches. Zoological Journal of the Linnean Society, 193(2), 381–387.
https://doi.org/10.1093/zoolinnean/zlab072
Faith, D. P. (2019). EDGE of existence and phylogenetic diversity. Animal Conservation, 22(6),
537–538. https://doi.org/10.1111/acv.12552
Garnett, S. T., & Christidis, L. (2007). Implications of changing species definitions for
conservation
purposes.
Bird
Conservation
International,
17(3),
187–195.
https://doi.org/10.1017/S0959270907000809
72

Garnett, S. T., & Christidis, L. (2017). Taxonomy anarchy hampers conservation. Nature News,
546(7656), 25.
Garnett, S. T., Christidis, L., Conix, S., Costello, M. J., Zachos, F. E., Bánki, O. S., Bao, Y.,
Barik, S. K., Buckeridge, J. S., Hobern, D., Lien, A., Montgomery, N., Nikolaeva, S., Pyle, R.
L., Thomson, S. A., van Dijk, P. P., Whalen, A., Zhang, Z.-Q., & Thiele, K. R. (2020).
Principles for creating a single authoritative list of the world’s species. PLOS Biology, 18(7),
e3000736. https://doi.org/10.1371/journal.pbio.3000736
Gasper, A. L. de. (2016). The fern family Blechnaceae: old and new genera re-evaluated, using
molecular data. Universidade Federal de Minas Gerais.
Gill,
Donsker,
&
Rasmussen.
https://www.worldbirdnames.org/new/

(2024).

IOC

World

Bird

List.

Gill, F., Donsker, D., & Rasmussen, P. (2023). IOC World Bird List (v13.1).
https://doi.org/10.14344/IOC.ML.13.1.
Harris, D. J., & Froufe, E. (2005). Taxonomic inflation: species concept or historical
geopolitical bias? Trends in Ecology & Evolution, 20(1), 6–7.
Hey, J., Waples, R. S., Arnold, M. L., Butlin, R. K., & Harrison, R. G. (2003). Understanding
and confronting species uncertainty in biology and conservation. Trends in Ecology &
Evolution, 18(11), 597–603. https://doi.org/10.1016/j.tree.2003.08.014
Hillis, D. M., Chambers, E. A., & Devitt, T. J. (2021). Contemporary Methods and Evidence for
Species Delimitation. Ichthyology & Herpetology, 109(3). https://doi.org/10.1643/h2021082
Hobern, D., Barik, S. K., Christidis, L., T Garnett, S., Kirk, P., Orrell, T. M., Pape, T., Pyle, R.
L., Thiele, K. R., & Zachos, F. E. (2021). Towards a global list of accepted species VI: The
Catalogue of Life checklist. Organisms Diversity & Evolution, 1–14.
Hortal, J., Bello, F. de, Diniz-Filho, J. A. F., Lewinsohn, T. M., Lobo, J. M., & Ladle, R. J.
(2015). Seven shortfalls that beset large-scale knowledge of biodiversity. Annu. Rev. Ecol.
Evol. Syst., 46, 523–552. https://doi.org/10.1146/annurev-ecolsys-112414-054400
Irham, M., Prawiradilaga, D. M., Menner, J. K., O’Connell, D. P., Kelly, D. J., Analuddin, K.,
Karya, A., Meads, M., Marples, N. M., & Rheindt, F. E. (2023). A distinct new species of
Zosterops white‐eye from the Sulawesi region, Indonesia. Ibis, 165(3), 808–816.
https://doi.org/10.1111/ibi.13148
Isaac, N. J. B., Mallet, J., & Mace, G. M. (2004). Taxonomic inflation: its influence on
macroecology and conservation. Trends in Ecology & Evolution, 19(9), 464–469.
Jarić, I., Correia, R. A., Roberts, D. L., Gessner, J., Meinard, Y., & Courchamp, F. (2019). On
the overlap between scientific and societal taxonomic attentions — Insights for conservation.
Science
of
The
Total
Environment,
648,
772–778.
https://doi.org/10.1016/j.scitotenv.2018.08.198
Jetz, W., Thomas, G. H., Joy, J. B., Hartmann, K., & Mooers, A. O. (2012). The global diversity
of birds in space and time. Nature, 491(7424), 444–448. https://doi.org/10.1038/nature11631
Kitchener, A. C., Hoffmann, M., Yamaguchi, N., Breitenmoser-Würsten, C., & Wilting, A.
(2022). A system for designating taxonomic certainty in mammals and other taxa. Mammalian
Biology, 1–11.
73

Ladle, R. J., & Hortal, J. (2013). Mapping species distributions: living with uncertainty.
Frontiers of Biogeography, 5(1). https://doi.org/10.21425/F55112942
Ladle, R. J., & Whittaker, R. J. (2011). Conservation Biogeography. In Conservation
Biogeography. https://doi.org/10.1002/9781444390001
Lepage, J., & Warnier, J. (2014). The Peters’ Check-list of the Birds of the World (1931-1987)
Database. https://avibase.bsc-eoc.org/peterschecklist.jsp
Lessa, T., Alves-Martins, F., Martinez-Arribas, J., Correia, R. A., Mendelsohn, J., Chimbioputo
Fabiano, E., Angombe, S. T., Malhado, A. C. M., & Ladle, R. J. (2024). Quantifying spatial
ignorance in the effort to collect terrestrial fauna in Namibia, Africa. Ecological Indicators, 158,
111490. https://doi.org/10.1016/j.ecolind.2023.111490
Lien, A. M., Conix, S., Zachos, F. E., Christidis, L., van Dijk, P. P., Bánki, O. S., Barik, S. K.,
Buckeridge, J. S., Costello, M. J., & Hobern, D. (2021). Towards a global list of accepted
species IV: Overcoming fragmentation in the governance of taxonomic lists. Organisms
Diversity & Evolution, 21(4), 645–655.
Mandiwana-Neudani, T. G., Crowe, T. M., Little, R. M., & Bowie, R. C. (2021). The
importance of adopting an integrative taxonomy framework in species delimitation: Response to
Hunter et al. (2021). Ostrich, 92(2), 162–167. https://doi.org/10.2989/00306525.2021.1898170
May, R. M. (1988). How Many Species Are There on Earth? Science, 241(4872), 1441–1449.
https://doi.org/10.1126/science.241.4872.1441
Mayr, E. (1946). The number of species of birds. The Auk, 63(1), 64–69.
McClure, C. J. W., Lepage, D., Dunn, L., Anderson, D. L., Schulwitz, S. E., Camacho, L.,
Robinson, B. W., Christidis, L., Schulenberg, T. S., Iliff, M. J., Rasmussen, P. C., & Johnson, J.
(2020). Towards reconciliation of the four world bird lists: hotspots of disagreement in
taxonomy of raptors. Proceedings of the Royal Society B: Biological Sciences, 287(1929).
https://doi.org/10.1098/rspb.2020.0683
Melo, M., Freitas, B., Verbelen, P., da Costa, S. R., Pereira, H., Fuchs, J., Sangster, G., Correia,
M. N., de Lima, R. F., & Crottini, A. (2022). A new species of scops-owl (Aves, Strigiformes,
Strigidae, Otus) from Príncipe Island (Gulf of Guinea, Africa) and novel insights into the
systematic
affinities
within
Otus.
ZooKeys,
1126,
1–54.
https://doi.org/10.3897/zookeys.1126.87635
Meyer, C. F. J., Aguiar, L. M. S., Aguirre, L. F., Baumgarten, J., Clarke, F. M., Cosson, J.-F.,
Estrada Villegas, S., Fahr, J., Faria, D., Furey, N., Weise, C. D., & Kalko, E. K. V. (2015).
Species undersampling in tropical bat surveys: Effects on emerging biodiversity patterns.
Journal of Animal Ecology, 84(1), 113–123. https://doi.org/10.1111/1365-2656.12261
Moura, M. R., & Jetz, W. (2021). Shortfalls and opportunities in terrestrial vertebrate species
discovery. Nature Ecology & Evolution, 5(5), 631–639. https://doi.org/10.1038/s41559-02101411-5
Navarro-Sigüenza, A. G., & Peterson, A. T. (2004). An alternative species taxonomy of the
birds of Mexico. Biota Neotropica, 4(2), 1–32. https://doi.org/10.1590/s167606032004000200013
Neate-Clegg, M. H. C., Blount, J. D., & Şekercioğlu, Ç. H. (2021). Ecological and
biogeographical predictors of taxonomic discord across the world’s birds. Global Ecology and
Biogeography, 30(6), 1258–1270. https://doi.org/10.1111/geb.13300
74

Orr, M. C., Feijó, A., Chesters, D., Vogler, A. P., Bossert, S., Ferrari, R. R., Costello, M. J.,
Hughes, A. C., Krogmann, L., Ascher, J. S., Zhou, X., Li, D.-Z., Bai, M., Chen, J., Ge, D., Luo,
A., Qiao, G., Williams, P. H., Zhang, A., … Zhu, C.-D. (2022). Six steps for building a
technological knowledge base for future taxonomic work. National Science Review, 9(12).
https://doi.org/10.1093/nsr/nwac284
Pante, E., Puillandre, N., Viricel, A., Arnaud-Haond, S., Aurelle, D., Castelin, M., Chenuil, A.,
Destombe, C., Forcioli, D., Valero, M., Viard, F., & Samadi, S. (2014). Species are hypotheses:
avoid connectivity assessments based on pillars of sand. Molecular Ecology, 24(3), 525–544.
https://doi.org/https://doi.org/10.1111/mec.13048
Parsons, D. J., Pelletierc, T. A., Wieringa, J. G., Ducketta, D. J., & Carstens, B. C. (2022).
Analysis of biodiversity data suggests that mammal species are hidden in predictable places.
Pnas, 119(14), 1–7. https://doi.org/10.1073/pnas.2103400119
Peterson, A. T., & Navarro-Sigüenza, A. G. (1999). Alternative species concepts as bases for
determining priority conservation areas. Conservation Biology, 13, 427–431.
Pyle, R. L., Barik, S. K., Christidis, L., Conix, S., Costello, M. J., van Dijk, P. P., Garnett, S. T.,
Hobern, D., Kirk, P. M., & Lien, A. M. (2021). Towards a global list of accepted species V. The
devil is in the detail. Organisms Diversity & Evolution, 1–19.
Rannala, B. (2015). The art and science of species delimitation. Current Zoology, 61(5), 846–
853. https://doi.org/10.1093/czoolo/61.5.846
Rheindt, F. E., Bouchard, P., Pyle, R. L., Welter-Schultes, F., Aescht, E., Ahyong, S. T.,
Ballerio, A., Bourgoin, T., Ceríaco, L. M. P., Dmitriev, D., Evenhuis, N., Grygier, M. J.,
Harvey, M. S., Kottelat, M., Kluge, N., Krell, F.-T., Kojima, J., Kullander, S. O., Lucinda, P.,
… Pape, T. (2023). Tightening the requirements for species diagnoses would help integrate
DNA-based descriptions in taxonomic practice. PLOS Biology, 21(8), e3002251.
https://doi.org/10.1371/journal.pbio.3002251
Riddle, B. R., Ladle, R. J., Lourie, S. A., & Whittaker, R. J. (2011). Basic Biogeography:
Estimating Biodiversity and Mapping Nature. In Conservation Biogeography.
https://doi.org/10.1002/9781444390001.ch4
Sangster, G. (2009). Increasing numbers of bird species result from taxonomic progress, not
taxonomic inflation. Proceedings of the Royal Society B: Biological Sciences, 276(1670),
3185–3191.
Stropp, J., Ladle, R. J., Emilio, T., Lessa, T., & Hortal, J. (2022). Taxonomic uncertainty and
the challenge of estimating global species richness. Journal of Biogeography, 49(9), 1654–1656.
https://doi.org/10.1111/jbi.14463
Stropp, J., Umbelino, B., Correia, R. A., Campos-Silva, J. V, Ladle, R. J., & Malhado, A. C. M.
(2020). The ghosts of forests past and future: deforestation and botanical sampling in the
Brazilian
Amazon.
Ecography,
43(7),
979–989.
https://doi.org/https://doi.org/10.1111/ecog.05026
Tessarolo, G., Ladle, R., Lobo, J. M., Rangel, T. F., & Hortal, J. (2021). Using maps of
biogeographical ignorance to reveal the uncertainty in distributional data hidden in species
distribution models. Ecography, 44(12), 1743–1755. https://doi.org/10.1111/ecog.05793
Thomson, S. A., Pyle, R. L., Ahyong, S. T., Alonso-Zarazaga, M., Ammirati, J., Araya, J. F.,
Ascher, J. S., Audisio, T. L., Azevedo-Santos, V. M., Bailly, N., Baker, W. J., Balke, M.,
75

Barclay, M. V. L., Barrett, R. L., Benine, R. C., Bickerstaff, J. R. M., Bouchard, P., Bour, R.,
Bourgoin, T., … Zhou, H.-Z. (2018). Taxonomy based on science is necessary for global
conservation. PLOS Biology, 16(3), e2005075. https://doi.org/10.1371/journal.pbio.2005075
Thomson, S. A., Thiele, K., Conix, S., Christidis, L., Costello, M. J., Hobern, D., Nikolaeva, S.,
Pyle, R. L., van Dijk, P. P., & Weaver, H. (2021). Towards a global list of accepted species II.
Consequences of inadequate taxonomic list governance. Organisms Diversity & Evolution,
21(4), 623–630.
Tobias, J. A., Sheard, C., Pigot, A. L., Devenish, A. J. M., Yang, J., Sayol, F., Neate-Clegg, M.
H. C., Alioravainen, N., Weeks, T. L., Barber, R. A., Thomas, G. H., & Schleuning, M. (2022).
AVONET: morphological, ecological and geographical data for all birds. Ecology Letters,
25(3), 581–597. https://doi.org/10.1111/ele.13898
Wägele, H., Klussmann-Kolb, A., Kuhlmann, M., Haszprunar, G., Lindberg, D., Koch, A., &
Wägele, J. W. (2011). The taxonomist - an endangered race. A practical proposal for its
survival. Frontiers in Zoology, 8(1), 25. https://doi.org/10.1186/1742-9994-8-25
Wheeler, Q. (2020). A taxonomic renaissance in three acts. Megataxa, 1(1), 4–8.
Winker, K., & Haig, S. M. (2010). Avian subspecies. In American Ornithologists’ Union (Ed.),
Ornithological Monographs (67th ed.).
Winker, K., & Rasmussen, P. C. (2021). Species limits and taxonomy in birds. Ornithology,
138(2). https://doi.org/10.1093/ornithology/ukab017
Yarwood, M. R., Weston, M. A., & Symonds, M. R. E. (2019). Biological determinants of
research effort on Australian birds: a comparative analysis. Emu - Austral Ornithology, 119(1),
38–44. https://doi.org/10.1080/01584197.2018.1501274
Zachos, F. E. (2016). Species Concepts in Biology. Springer International Publishing.
https://doi.org/10.1007/978-3-319-44966-1
Zachos, F. E. (2018). (New) Species concepts, species delimitation and the inherent limitations
of taxonomy. Journal of Genetics, 97(4), 811–815.

76

Capítulo 3
Revista: Ecological Indicators
Status: Publicado (doi.org/10.1016/j.ecolind.2023.111490)
Quantifying spatial ignorance in the effort to collect terrestrial fauna
in Namibia, Africa
Thainá Lessaa,b*,§, Fernanda Alves-Martinsb,c,§, Javier Martinez-Arribasb,c, Ricardo A.
Correiad,e,f, John Mendelsohng, Ezequiel Chimbioputo Fabianoh, Simon T. Angombei,
Ana C.M. Malhadoa,b,c, Richard J. Ladlea,b,c
a

Institute of Biological and Health Sciences, Federal University of Alagoas, 57072-970

Maceió, Alagoas, Brazil.
b

CIBIO-InBIO, Research Centre in Biodiversity and Genetic Resources, University of

Porto, Campus de Vairão, 4485-661 Vairão, Portugal.
c

BIOPOLIS Program in Genomics, Biodiversity and Land Planning, CIBIO, Campus de

Vairão, 4485-661 Vairão, Portugal
d

Helsinki Lab of Interdisciplinary Conservation Science (HELICS), Department of

Geosciences and Geography, University of Helsinki, 00014 Helsinki, Finland.
e

Helsinki Institute of Sustainability Science (HELSUS), University of Helsinki, 00014

Helsinki, Finland.
f

DBIO & CESAM, Centre for Environmental and Marine Studies, University of Aveiro,

3810-193 Aveiro, Portugal.
g

Ongava Research Centre, Ombika, Namibia.

h

Department of Wildlife Management & Tourism Studies, Katima Mulilo Campus,

University of Namibia, Windhoek, Namibia.
i

School of Agriculture and Fisheries Sciences, Neudamm Campus, University of

Namibia, Windhoek, Namibia
*Corresponding author: Thainá Lessa and Fernanda Alves-Martins.
Emails: thainalessaps@gmail.com, ferfealvesmartins@gmail.com.
§ These authors contributed equally to this work.

77

Abstract
Effective conservation efforts and predictions of future impacts on biodiversity depend
heavily on publicly available information about species distributions. However, data on
species distributions is often patchy, especially in many countries of the Global South
where resources for biological surveys have been historically limited. In this study, we
use biodiversity ignorance scores to quantify and visualize gaps and biases in
biodiversity data for Namibia, with a focus on five terrestrial taxa at a spatial scale of 10
x 10 km. We model the relationship between ignorance scores and socio-geographical
variables using generalized additive models for location, scale and shape (GAMLSS).
Our findings demonstrate that despite a high volume of occurrence records available on
the Global Biodiversity Information Facility (GBIF), publicly available knowledge of
Namibia's terrestrial biodiversity remains very limited, with large areas contributing few
or no records for key taxa. The exception is birds that have benefitted from a massive
influx of data from the citizen science platform eBird. Our study also highlights the
importance of citizen science initiatives for biodiversity knowledge and reinforces the
usefulness of ignorance scores as a simple intuitive indicator of the relative availability
and distribution of species occurrence records. However, further research, biological
surveys, and renewed efforts to make existing data held by museums and other
institutions widely available are still necessary to enhance biodiversity data coverage in
countries with patchy data.
Key-words: GBIF, occurrence records, survey effort, ignorance scores, species
distributions, vertebrates, Southern Africa.

78

Introduction
The development of new surveying tools and national and international biodiversity
information systems is making existing species records available to researchers
worldwide via the internet (Hedrick et al., 2020). The most important of these
endeavours is the Global Biodiversity Information Facility (GBIF) that was started in
2001 (Edwards, 2004). The GBIF takes advantage of long-term, coordinated and
ongoing efforts to digitize specimens from world’s natural history collections (Gaijy et
al., 2013; Nelson & Ellis, 2019) and, more recently, from some citizen science
databases (Chandler et al., 2017). By most measures the GBIF has been a remarkable
success, and currently hosts over two and half billion species occurrence records from
over two thousand institutions and open access data repositories (GBIF, 2023).
However, even enormous databases such as the GBIF are incomplete and uncertain,
with a considerable amount of biodiversity data subject to errors, often due to the low
accuracy, low precision and lack of standardization from multiple data sources (Barve
& Otegui, 2016; Cobos et al., 2018; D’Antraccoli et al., 2022; Ladle & Hortal, 2013).
Accepting that biases and gaps in biodiversity data often cannot be avoided,
especially in the more unevenly sampled countries and regions of the world (Danovaro
et al., 2010; Hopkins, 2019; Lessa et al., 2019), it becomes important to quantify and
understand the limits of our biodiversity knowledge. There are several ways to evaluate
biodiversity knowledge gaps and data quality. One recent proposal is through the
creation of ‘maps of biogeographical ignorance’ (MoBIs) that distinguish data quality
degree and intensively sampled areas from poorly sampled ones (Rocchini et al., 2011;
Tessarolo et al., 2021). MoBIs are typically based on a combination of: i) completeness
of species inventories for defined sampling units (e.g., Stropp et al., 2016); ii) estimates
of taxonomic quality, and; iii) temporal and spatial decay in data (Ladle & Hortal, 2013;
Tessarolo et al., 2021). Unfortunately, MOBIs are not appropriate to use in data-poor
regions because they are very sensitive to low record numbers and to non-natural
relative abundances of species in natural history collections (Meyer et al., 2016; Steege
et al., 2011; Stropp et al., 2016); for example, the overrepresentation of rare species in
museum collections relative to their true abundances (Gotelli et al., 2021).
A simpler alternative to MoBIs is to quantify and map the absence of data (i.e.,
ignorance) in biodiversity databases through the ‘ignorance scores’ approach (Ruete,
2015). These are useful to rapidly assess and visualize biases and shortfalls related with
taxonomy, temporal and spatial data. Ignorance scores can also be used to characterize
79

the degree of biodiversity knowledge based on the effort (or weakness of it) to record
species occurrences (Correia et al., 2019; Mair & Ruete, 2016; Ruete, 2015). The
approach has the added advantage of being simple to calculate, does not involve
prediction or estimation of the total number of species in a given area, has a very limited
number of assumptions, and relies solely on raw data of species occurrences, i.e.,
presence-only data (Correia et al., 2019). The score provides information on recording
coverage and reliability that can be used to measure the spatial distribution of recording
effort across a study area, and thus to identify undersampled and priorities areas for data
collection (Mair & Ruete, 2016).
Multiple factors have been identified as potentially influencing recording effort.
For example, reasons associated with how accessible and/or practical it is to sample a
given area, such as road density, human population density, or proximity to universities
(Meyer et al., 2015; Sastre & Lobo, 2009), and public and/or scientific interest are
known to positively influence the site selection for recording biodiversity (Millar et al.,
2019; Oliveira et al., 2016). Collectors (biodiversity researchers and citizen scientists)
often prefer to sample sites perceived as being poorly studied, ecologically unique, more
diverse or well-preserved, such as formally protected areas (PA) or sites with pristine
native vegetation (Boakes et al., 2010; Rocha‐Ortega et al., 2021; Yang et al., 2014).
Additionally, collectors frequently prefer to sample areas near research centres (Ribeiro
et al., 2016; Carvalho et al., 2023), which are typically located in economically more
developed regions (Meyer, 2016). Species distribution data thus tend to vary more
among political than ecological units, reflecting historical patterns of collecting,
collating and digitalizing biogeographical data (Hortal et al., 2015; Stropp et al., 2016).
In unevenly sampled regions this can lead to maps of species richness that closely
resemble maps of survey effort (Hortal et al., 2015), a pattern that is particularly striking
in sub-Saharan Africa (Stropp et al., 2016).
Namibia is a large, arid southwest African country with high levels of endemism
and low human population density (Atlas of Namibia Team, 2022; Simmons et al.,
1998). It has a strong system of protected areas, but Namibia's species occurrence
records are very patchy on publicly available platforms such as GBIF (GBIF, 2023).
These characteristics make the country an ideal political unit to evaluate the spatial
patterns of biodiversity ignorance through ignorance scores (Lessa et al., 2019).
Therefore, we applied the ignorance score approach to evaluate variation in species
occurrence records available from GBIF across multiple terrestrial taxa for Namibia.
80

Specifically, we used species records collected from GBIF to: i) characterize temporal
and taxonomic biases in recording efforts; ii) evaluate and map spatial shortfalls in
recent recording effort, iii) analyse the influence of multiple socio-geographical
variables on the distribution of recent recording effort, and iv) discuss the usability of
ignorance score approach to evaluate quality in publicly available biodiversity data.
Methods
Study area
Namibia is a southwest African country with a terrestrial area of approximately 824,000
km2 (Figure 1). Its geomorphology is dominated by the great escarpment along the
western side of the country, forming a transition between the narrow coastal desert and
a flat inland plateau dominated by aeolian sand. Namibia is the most arid sub-Saharan
country (Gargallo, 2020), with the Namib desert in the southwest of the country
receiving an annual average precipitation of less than 50 mm. Moreover, rainfall is very
variable, mostly falling over short, intense periods (Atlas of Namibia Team, 2022).

Figure 1: Map of Namibia highlighting socio-geographical variables used in our analysis, for
example, roads, protected areas, vegetation cover and density of people.

There are few permanent rivers; the Kunene and Okavango Rivers form the
northern border with Angola, the Kwando, Linyanti, Zambezi and Chobe Rivers form
the borders with Botswana and Zambia, and the Orange River borders South Africa in
the south. The vegetation of Namibia can be broadly classified in deserts (16% of the
81

country), savannas (64%) and woodlands (20%) (Giess, 1971) with both summer and
winter rainfall zones. Over 70% of the country is classified as arid or semi-arid
(Simmons et al., 1998). The great variability in rainfall means that the amount of
standing herbaceous vegetation varies considerably from year to year (Wardell-Johnson,
2000).
Species occurrence records and filtering processes
We collected all species occurrence records (hereafter just ‘records’) available for
Namibia in GBIF (https://www.gbif.org/). We chose to collect data from GBIF because
it has an international mandate to compile global species records and is one of the most
commonly used sources of data for biodiversity studies globally. We first collected data
from records of Namibia (1,656,016 records, GBIF, 2021). The following methods were
used to exclude records: i) suspicious geographical coordinates - these are records with
geographical coordinates assigned to the centroid of a municipality, state, country or
falling in the ocean; ii) invalid, unlikely, mismatched or absent collection dates; iii)
absent taxonomic identification at species level, or; iv) uncertain taxonomic data at
species level - taxonomic data that does not match any known species or where matches
can only be obtained through fuzzy matching. Records were collected with the ‘rgbif’
library for R software (Chamberlain & Boettiger, 2017). The precision of geographical
coordinates was examined with the ‘CoordinateCleaner’ library for R software (Zizka et
al., 2019). Given that GBIF includes datasets with varying coordinate precision, we
accepted geographical coordinates with varying levels of precision, allowing for a
certain degree of rounding. However, these coordinates were accepted only if they kept
a precision greater than the spatial unit used in the study, which was the 10 km cell
(Zizka et al., 2020). The validation of scientific names was performed with the ‘taxize’
library for R software (Chamberlain et al., 2020) and manually. The cleaning process
returned 1,139,786 records. Then, we filtered only records on five reference taxonomic
groups that would be evaluated (as described below), which resulted in a dataset with
1,075,916 records collected from 1783 to 2021 (full dataset).
We arranged records into subcategories of reference taxonomic groups, which
refer to a group of species that can be studied or collected across similar methodologies.
All species belonging to a reference taxonomic group probably share the record bias
analogously. In these cases it is standard practice to use occurrence counts of species
from that taxonomic group as a substitute for its recording effort (Phillips et al., 2009;
82

Ponder et al., 2001). The assumption underlying this statement is that the absence of
records for any species from a reference taxonomic group (e.g. mammals) in a particular
area is likely due to a lack of a specialist, rather than the total absence of the reference
group in the area. Similarly, if there are many records of a reference taxonomic group in
a given place, it is likely that the lack of records of a particular species in that place is
due to true absence (Phillips et al., 2009; Ruete, 2015). We considered five taxonomic
groups for calculating ignorance scores: birds (Aves), mammals (Mammalia),
amphibians (Amphibia), reptiles (Reptilia) and insects (Insecta).
After the data cleaning process, we carried out a temporal filtering process
keeping only records collected from 2000 onwards for making ignorance scores and
maps, and spatial analysis. The chosen time window reduces the probability of changes
in collection behaviours, thereby minimizing recording biases (Ponder et al., 2001;
Ruete, 2015) and ensuring that the period of time the records were collected is
congruent with the socio-geographic variables used to explore recording biases (details
in section 2.4). The temporal filtering dataset (from 2000-2021; hereafter, recent
dataset) returned 1,010,800 records which were analysed to create ignorance scores and
maps (section 2.3) and explore factors influencing the spatial distribution of records
(section 2.4).
Ignorance scores and maps
We calculated ignorance scores for each reference taxonomic group over the entire
Namibian territory by defining recording units (SUs) of 10 x 10 km, using the recent
dataset (data from 2000-2021). For this, we generated a raster grid using an equal-area
(Eckert IV) projection, which returns 8,567 grid cells. We chose square grid because it
is the most commonly used polygon shape in spatial analysis by ecologists, it is simple
for calculations, transformations and comparisons, and is frequently used in Geographic
Information Systems rasters (Birch et al., 2007). The 10 x 10 km spatial resolution was
chosen because it was considered an adequate size to be reasonably sampled during a
recording visit (Correia et al., 2019). We then converted the grid to WGS84 to match
species records projection. Ignorance scores were calculated using the ‘LogNormalization approach’ suggested by Ruete (2015) – ignorance is equal to one minus
the normalization of the natural logarithm of the data – and defined by the following
equation: Ignorance score = 1 – (ln(Ni +1)/ln(Nm+1)).

83

Where Ni is the number of records in a grid cell i and Nm, the maximum number
of records in the cell with the highest number of records. For example, in the case of
Birds the highest number of records (N = 29,914) was found in a cell near Windhoek.
Therefore, the 'Log-Normalization approach' considered the maximum value of 29,914
when calculating the ignorance score for birds. The ‘Log-Normalization approach’
transforms records counts into a 0-1 scale of ignorance, with a score of one indicating
complete ignorance, i.e., no single record available for the cell, and a score of 0
indicating the best available knowledge, i.e., the maximum number of records (Nm).
This approach is the most suitable when there are large differences in the minimum and
maximum number of records for a given reference taxonomic group, which is our case
(Birds = 1-29,914; Mammals = 1-594; Reptiles = 1-79; Amphibians = 1-21; Insects = 1540) and allows comparisons among the distinct reference taxonomic groups.
Environmental and socio-geographical drivers of species recording effort
We collected data on five socio-geographical variables that may drive the spatial
distribution of recording effort based on perceptions of site accessibility or biological
value: 1) road density; 2) human population density; 3) minimum distance to a
university; 4) minimum distance to a protected area; and 5) average vegetation cover.
Road density was estimated as the total length of roads (in km) in an area of 100 km2
(10 x 10 km grid cells) covering the Namibian territory based on data from the
OpenStreetMap database (see Correia et al. (2019) for a similar approach). Human
population estimates for the years 2000, 2005, 2010, 2015, and 2020 were obtained at 1
km resolution from the Center for International Earth Science Information Network –
CIESIN – Columbia University (2018), and aggregated for the grid cells resolution (10
x 10 km) by summing cells’ values, so that population density refers to the total count
of people in cells of 100 km2. Vegetation cover at 2000 was obtained at 30 meters’
resolution from (Hansen et al., 2013) and aggregated for the grid cells resolution by the
mean value of cells. Minimum distance to universities was calculated for each grid cell
based on the location of universities and other higher education institutions (e.g.,
colleges) inside the country. The geographical location of each higher education
institution was obtained from OpenStreetMap database. Grid cells containing at least
one higher education institution were assigned a distance of zero. For grid cells without
any higher education institution, the distance of the cell centroid to the nearest higher
education institution was estimated. Minimum distance to Protected Areas (PAs) was
84

calculated for each cell based on the location of PAs in the region. Maps of PAs were
obtained from the World Database on Protected Areas on November 2021 and include
national parks, private reserves, communal conservancies among other categories of
protected areas (available from https://www.protectedplanet.net). Cells covered by
protected areas were assigned a distance of zero, otherwise the distance from the cell
centroid to the nearest boundary of a protected area was calculated.
Spatial analyses were performed on QGIS 3.20. We used Spearman’s correlation
to assess pairwise correlation among variables and observed a weak correlation (rs < 0.3
for all variable pairs), except for population density and vegetation cover, which
exhibited a correlation coefficient of 0.53. The exclusion of population density during
the model selection process (see below and Supplementary Material 1) mitigates any
multicollinearity issues arising from this.
Data analysis
Initially we used the full dataset (cleaned records from 1783-2021) to characterize
temporal and taxonomic biases in recording efforts. To do this, we created bar and
spider graphs using R software. Afterwards, we used the recent dataset (cleaned records
from 2000-2021) to create ignorance scores and ignorance maps for the five reference
taxonomic groups, and to perform statistical analysis. When exploring ignorance scores
and maps, we found a large proportion of grid cells that had ignorance scores of 1 (i.e.,
without any record). Based on this evaluation, we used Generalized Additive Models
for Location, Scale and Shape (GAMLSS) (Rigby & Stasinopoulos, 2005) to explore
the relationship between ignorance scores and the multiple environmental and sociogeographical variables outlined for Namibia.
GAMLSS was chosen because our response variable, ignorance scores, follow a
one-inflated beta distribution (“BEINF1”; Ospina & Ferrari, 2010), with values ranging
between 0 and 1 (0<Ignorance score<=1) and containing a large proportion of ignorance
scores of 1. This distribution is suitable when there is an excess of ones in the data
compared to what would be expected from a standard beta distribution, and cannot be
modelled using a classical Generalized Linear Model (GLM) approach. In addition to
allowing the use of a wide range of statistical distributions, GAMLSS can deal with
heterogeneous, highly skewed and kurtotic data, such as the left-skewed distribution of
the ignorance scores. We converted ignorance scores equal to zero to 10e-06, as the logNormalization approach returns ignorance scores equal to zero for cells with the
85

maximum number of records, and the one-inflated beta distribution only accepts values
greater than zero.
GAMLSS models assume that the response variable is described by a density
function defined by up to 4 parameters (μ, σ, ν, τ) that determine its location μ (i.e.,
mean), scale σ (i.e., standard deviation) and shape (i.e., skewness ν and kurtosis τ)
(Stasinopoulos & Rigby, 2007). We examined the relationship between ignorance
scores and socio-geographical factors by assessing how these factors affect the location
(i.e., the mean), skewness and kurtosis (i.e., the shape of the relationship). To capture
non-linear relationships, we applied a smoothing function (P-splines). Finally, we used
a model selection approach based on Generalized Akaike Information Criterion (GAIC)
scores to select the most informative socio-geographical variables for each reference
taxa model. GAIC is an extension of AIC (Akaike Information Criterion), which takes
into account the additional complexity of GAMLSS models, which have more
parameters than traditional GLM models, and therefore include a higher penalty for the
number of parameters in the model. In general, the smaller the GAIC value, the better
the model fit (Stasinopoulos et al., 2017).
We ran GAMLSS models for the reference taxonomic groups. GAMLSS models
were calculated independently. All model results, including the relative explanatory
power of each model, are reported in Supplementary Material 2. Statistical analyses
were carried out in R statistical software 4.2.0 (R Team Core, 2017) using the ‘gamlss’
package (Rigby & Stasinopoulos, 2005). Models were implemented with the ‘gamlss’
function and pseudo R-squared values for each model were obtained with function
‘Rsq’ using option ‘Cragg Uhler’ (Stasinopoulos & Rigby, 2007).
Results
Based on records incorporated into GBIF, some clear temporal biases were observed in
recording effort over the nearly 240 years (the full temporal window - 1783-2021) of
recording biodiversity in Namibia. Specifically, there are very few records available
before the 1990s, representing only 3.2% of all data. The highest peaks of records
occurred from the 1990s onwards. The first peak occurred between 1993-2007 holding
14.8% of the full dataset, and the second peak between 2008-2019 with the highest
number of records (71.7%) - five times more recording effort than the initial period. The
year with the highest volume of records was 2019 (Figure 2, grey line).

86

The temporal biases in recording effort for each taxonomic group followed a
similar pattern to the overall dataset for birds (Figure 2, blue line). For mammals, the
largest influx of records into GBIF was after 2000s, however a significant influx of
records was noted in 1970s (red line). For reptiles and amphibians, the greatest
recording efforts were made between the 1970s and 1990s, with few records collected
and/or available after the 2000s (orange and purple line, respectively). Finally, the
pattern of data influx for insects was more uniform when compared to the other
taxonomic groups, showing peaks in the number of records collected in the 1920s,
1970s and 2000s (green line).

Figure 2: Historical progression of the number of occurrence records for Namibia's biodiversity
publicly available on GBIF platform (full dataset). Number of records of all taxa (grey line) and
separately, according to the fauna silhouette.

Despite the relatively large volume of data on Namibia’s biodiversity available
in the GBIF from 1783-2021, our analysis still revealed strong biases in terms of
taxonomic groups and in the characteristics of the records. Approximately 94% (n =
1,011,197) of records in the full dataset refers to birds, and 99.6% of these records were
from human observations rather than specimens. Birds had the lowest rate (2.5%) of
data loss after the temporal filtering process (2000-2021; recent dataset) (Figure 3). The
second most representative reference taxonomic group was insects, with 26,707 records
in the full dataset, though 86.6% of these records came from preserved specimens rather
than observations. However, 65.4% of records in the full dataset were lost after the
temporal filtering process (2000-2021). The records of mammals represented only 1.8%
87

of the full database, and after the year 2000 there was a decrease of 44.5% in the
number of records. About 55% of mammals’ records came from observations. The most
critical shortfall was in Herpetofauna records. With 1,540 records, amphibians showed
the lowest number of records in the full dataset (0.14%), and the highest rate (85%) of
record loss after 2000. Furthermore the low number of records of amphibians implies
that most species in our dataset are represented by only one or few records (i.e.,
singletons, doubletons, etc.). Reptiles showed the second lowest number of records in
the full dataset (16,916 records) and 76% was lost after 2000s. Over 85% of
herpetofauna records are based on preserved specimens (Figure 3).

Figure 3: Bar graphic indicates the number of records in full dataset (1783-2021) and
recent dataset (2000-2021). Radar chart illustrates the percentages of GBIF’s basis of
records using full dataset, with HO=Human Observation and PS=Preserved Specimen.
Notable gaps and biases for all reference taxonomic groups were observed in
spatial distribution of ignorance scores in Namibia. A temporal decline in ignorance
scores was noted when accumulating records (see temporal figure in Supplementary
material 3). In more populated areas, such as the capital Windhoek, the coastal cities
Swakopmund and Walvis Bay, the central cities Otjiwarongo and Okahandja, Rundu
(northeast) and Keetmanshoop (southern), were characterized by high recording effort,
and consequently, low ignorance scores. The coastal zone, where the Namib Desert is
located, had low ignorance scores for mammals, birds and insects. The Succulent Karoo
region showed low ignorance only for birds and insects. For almost all taxonomic
groups (except birds), the eastern portion of Namibia – a scrub savanna region of

88

aeolian sands bordering Botswana and South Africa – showed lower recording efforts
and thus, high rates of ignorance scores (Figure 4).

Figure 4: Spatial distribution of GBIF’s records and ignorance scores for Namibia’s birds,
mammals, reptiles, amphibians and insects. Maps were calculated from recent dataset (20002021). Ignorance maps represent a gradient of ignorance scores - from cells with high ignorance
scores (purple tons) to cells with low ignorance scores (yellow tons). Histograms represent the
frequency of cells according to ignorance scores gradient. Silhouettes refer to taxonomic groups.

Analysing only the recent dataset (from 2000-2021), large areas of the country
were underrepresented, even for birds the group with the highest number of records in
GBIF (Figure 4). The ignorance maps for birds revealed that 52.4% of grid cells had no
single record (ignorance score = 1), 17.3% had between 1 to 10 records (ignorance
score = 0.93–0.76) and 5.7% had between 50 to 100 records (ignorance score = 0.61–
0.55). Conversely, a small portion of the country was found to be overrepresented, i.e.,
0.17% of 10 x 10 km2 grid cells had between 30,419 to 10,405 records (ignorance score
= 0–0.1) and 1.78% of 10 x 10 km2 grid cells had between 9,185 to 1,003 records
(ignorance score = 0.11–0.33) (yellowest parts of the map in Figure 4). Spatial biases in
the distribution of records were more pronounced in the other reference taxonomic
groups. The percentage of grid cells without any records (ignorance score = 1) was 99%
to amphibians, 90.5% to reptiles, 90.2% to insects and 87.6% to mammals. In contrast,
89

the percentage of grid cells that had ignorance score ≤ 0.5, i.e., where there is a greater
recording effort, ranged between 0.1% (amphibians), 1% (insects), 1.1 % (mammals)
and 1.2% (reptiles) (Figure 4).
Unsurprisingly the socio-geographical variable that best drives recording effort
for all taxonomic groups analysed was road density, with an overall bias towards
recording specimens in more accessible areas (higher road density, lower ignorance
scores; Table 1). The distance to universities was also a significant predictor for bird
and amphibian records. Ignorance scores exhibited an increase as the distance from
universities increased, suggesting a tendency to record species in close proximity of
these institutions. Maps of ignorance show a tendency for mammal, reptile and
amphibian records to be collected in protected areas, such as national parks Etosha,
Bwabwata, Waterberg Plateau, Skeleton Coast, Tsau and Namib-Naukluft. The eastern
portion of Namibia (scrub savanna region), which had higher ignorance scores, is not
covered by protected areas. Our statistical analyses validated these observations,
demonstrating that the distance from protected areas is a significant driver of recording
effort for these taxa. Ignorance scores increase with distance from PAs, indicating a
decrease in recording efforts for sites located far from protected areas. Finally, the
percentage of vegetation cover showed effect on recording effort only for birds, the
ignorance scores for this taxon were lower in areas with less vegetation coverage, since
much of Namibia’s vegetation is composed of savannah, dry woodlands and desert
(Table 1).
Table 1: Significant results of GAMLSS models exploring the association between ignorance
scores and environmental and socio-geographical factors for the five reference taxonomic
groups in Namibia (p < 0.05). The complete results, including non-significant associations, are
available in the Supplementary Material 2.
Variable

Road density

University distance
Protected Area Distance
Forest Cover

Reference taxonomic
group
Amphibians
Birds
Insects
Mammals
Reptiles
Amphibians
Birds
Amphibians
Mammals
Reptiles
Birds

Coefficient
T value P
estimate
-0.666
-4.767
0.000
-1.030 -23.582
0.000
-0.356
-5.073
0.000
-0.518
-8.240
0.000
-0.448
-5.592
0.000
-0.002
-2.539
0.011
0.001
9.156
0.000
0.005
2.279
0.023
0.005
6.257
0.000
0.004
4.408
0.000
-0.041
-7.471
0.000
90

Discussion
Our main finding is that the volume of publicly (and thus widely) available digital
information about Namibia's mainland fauna on the GBIF is still very low for most taxa
and regions. Except for birds, all reference taxonomic groups evaluated here have
significant temporal and spatial data shortfalls, and most of the records that have so far
been added to GBIF were collected before 2000’s and are therefore subject to higher
rates of data degradation (Tessarolo et al., 2017). Interestingly, however, there were
peaks in collection and availability of records during the period when Namibia was
engaged in armed conflict for its independence (1960-1980). Although the amount of
GBIF records appears to be increasing rapidly from 1990 onwards, when Namibia
finally became independent, it is important to note that much of this increase is being
driven by the recent influx of information on birds from the eBird citizen science
platform (Bonney, 2021). eBird is also likely to be the main driver behind the decline in
GBIF records during the 2020 COVID-19 pandemic when international travel was
restricted and visits to national parks around the world fell precipitously (Hockings et
al., 2020; Souza et al., 2021).
That Namibia has a low number and coverage of biological records is perhaps
unsurprising given that it is the driest country in Sub-Saharan Africa (Simmons et al.,
1998) with all of the associated challenges that this poses for biological surveying and
collecting in arid, inhospitable environments with limited accessibility (Boakes et al.,
2010; Lessa et al., 2019). In a recent assessment of insects’ (Lepidoptera, Sphingidae)
inventory completeness in Sub-Saharan Africa, a large proportion of Namibia had
between 1-50 records in 200 x 200 km grid cells, and 5-30% of these were complete
(Ballesteros-Mejia et al., 2013). For plants, Namibia showed a low proportion of wellsampled areas and much of the data was missing, incipient and outdated (Stropp et al.,
2016). The persistence of more obsolete data compromises our understanding of the true
composition of biodiversity, making conservation actions potentially inaccurate and
inefficient (Escribano et al., 2016). Nevertheless, Namibia has a considerable need for
publicly available high quality biodiversity information compared to other arid and dry
regions – such as its neighbouring South Africa, which holds 30 times more records in
GBIF, including more recent and complete data (Stropp et al., 2016) – given the
enormous interest and economic importance of its wildlife industry (Schalkwyk et al.,
2010).

91

We found that road density, a proxy of accessibility, was most strongly related to
recording effort for all modelled taxonomic groups. This is a long-recognized bias for
records, both historical and contemporary, and is often referred to as ‘roadside’ bias or
the ‘roadside effect’ (Oliveira, et al., 2016; Petersen et al., 2021). A similar pattern was
observed for the location and density of passerine birds and hawkmoths records in subSaharan Africa, which had a higher effort in more accessible locations: close to roads,
railway lines, airports, rivers and cities (Reddy & Davalos, 2003; Ballesteros-Mejia et
al., 2013). The probable mechanism behind this bias is that observations are more
frequently made at short distances from roads and paths due to easier accessibility and
convenience for collectors/surveyors (Kadmon et al., 2004; Petersen et al., 2021; Sastre
& Lobo, 2009). This may be especially true in more inhospitable environments. Some
researchers have also observed a trend of more records in densely populated areas
(Luck, 2007), presumably for similar reasons. As Petersen et al., (2021) point out, the
major concern over this particular bias is that areas close to roadsides may not be
representative of the wider landscape, potentially leading to incorrect inferences about
biodiversity patterns (but see Revermann et al. 2017).
Perhaps our most surprising result was the lack of influence of population
density on ignorance scores of reference taxonomic groups as this variable was
excluded by the Generalized Akaike Information Criterion (GAIC). This finding can be
explained by Namibia’s very low population density. With over 2.6 million people (The
World Bank, 2022), Namibia is one of the most sparsely populated countries in Africa
(and the world). It has an average density of 2.5 persons per square kilometer (Wart et
al., 2015), except for urban centres such as Windhoek, Rundu, Walvis Bay and
Swakopmund, and certain densely populated rural areas in the central north and northeastern areas of the country (Figure 1). Our findings indicated that 99% of Namibia
territory has no amphibian records, 90.5% no reptiles, 90.2% no insects and 87.6% no
mammals. In a previous study in a similarly arid region, the variables human population
density and road density were spatially correlated (Oliveira et al., 2016; Correia et al.,
2019). However, although the area in question – Brazilian Caatinga – has a geographic
size similar to Namibia, it has almost ten times more inhabitants.
Our model also showed a relationship between recording effort and distance to
universities for birds and amphibians. Again, this can be interpreted as a form of
‘convenience bias’; a high proportion of individuals contributing data to the GBIF are
from the university sector (Correia et al., 2019; Liu et al., 2021) and, ceteris paribus,
92

they will be more likely to collect records from sites close to their place of work than
more distant sites. This behaviour is likely to have several underlying causes, including
the practical and financial burden of mounting research expeditions to more remote
areas, the increased likelihood of field stations and other research infrastructure closer
to the university, and the added scientific value of working on a site that has already
been partially or fully documented (dos Santos et al., 2015).
We found a clear tendency for records of all reference taxonomic groups to be
collected in protected areas, such as the National Parks of Etosha, Bwabwata,
Waterberg Plateau, Skeleton Coast, Tsau and Namib-Naukluft. Approximately 40% of
Namibia territory has some degree of protection (Corrigan et al., 2018). In an extremely
fragmented world, Namibia bucks the trend, connecting and protecting its areas in terms
of ecological and economic values through ecotourism. This observed tendency is not
surprising since we would expect both academics and amateur naturalists to take
advantage of the superior infrastructure and accessibility available in these areas. The
positive impact of protected areas on GBIF records has been noted previously (Correia
et al., 2019; Oliveira et al., 2016), including in Sub-Saharan Africa (Ballesteros-Mejia et
al., 2013), where several studies have shown that research sites tend to cluster close to
universities in areas with some form of protection (dos Santos et al., 2015; Lessa et al.,
2019). Despite the clear tendency of records to be associated with protected areas, we
found very low recording effort for all reference taxonomic groups in national parks in
the eastern portion of the country. This region has the lowest protected areas coverage
and should be prioritized in new field works and recording efforts, or if these records
already exist they should be made publicly available. Although our model did not reveal
a significant association between recording effort and bird records in protected areas,
these areas are still important for bird conservation according to the Atlas of Namibia
(2022). Increased research effort could lead to a higher number of records if researchers
make their data available through digital platforms such as GBIF.
Citizen science initiatives have played a significant recent role in mobilizing
biodiversity data to the GBIF. Half of all records shared via GBIF come from datasets
with significant volunteer contributions (Chandler et al., 2017).

This trend is

particularly notable in the case of Namibian birds and mammals, where a large amount
of records came from citizen science platforms such as eBird, Southern African Bird
Atlas Project 2 (SABAP2), the South African Bird Ringing Unit (SAFRING),
iNaturalist and Observation.org. These platforms are almost universally used by
93

amateur and professional wildlife watchers and photographers, resulting in a remarkable
increase of the information available in GBIF (Bonney, 2021). This effect is especially
notable in countries like Namibia, due to the large influx of international ecotourists.
Although for other reference taxonomic groups citizen science appears to be a less
common source of biodiversity data in Namibia. Specifically, amphibians and reptiles
present a high proportion of records based on specimens preserved in museum
collections.
Ignorance scores as a tool for visualizing biodiversity data needs
The concept of Ignorance Scores was introduced by Ruete (2015) and subsequently
applied by Correia et al., (2019) for the semi-arid Caatinga biome of northeast Brazil. In
contrast to alternative approaches proposed to evaluate data quality and completeness,
such as Inventory Completeness (Sousa-Baena et al., 2014; Stropp et al., 2016) and
MoBIs (Hortal et al., 2022; Tessarolo et al., 2021), Ignorance Scores stand apart by their
unique approach. By exclusively relying on the presence of data in their metrics,
without considering observed or expected species richness, ignorance scores can be
applied in areas with very few data. Specifically, key distinctions between these
approaches can be delineated as follows: i) Ignorance Score: bases on raw data, does not
require a minimum number of records, and there are no estimations of species richness;
ii) Completeness: draws upon species accumulation curves, requires a minimum number
of records and estimates the number of species; and iii) MoBIs: integrates several data
sources (completeness, temporal and spatial decay, and taxonomic quality) and requires
a minimum number of records.
The advantage of the Ignorance Score approach is that it provides “a simple and
intuitive indicator of species recording effort, allowing the assessment of taxonomic and
spatial biases present in the GBIF database” (Correia et al., 2019, p. 8). Furthermore, it
is ideal for regions or countries where there are large areas with few or no records where
it would be impossible to compute more sophisticated measures of recording
completeness based on species accumulation curves (e.g. Sousa-Baena et al., 2014). The
ignorance score approach is extremely flexible and can be easily computed and mapped
at different spatial scales and can be used, as in the current case study, to provide a rapid
visual indicator of areas in need of further recording effort (Correia et al., 2019; Ruete,
2015).

94

Our study clearly shows an urgent need of collection efforts and mobilization of
existing biodiversity data in the eastern and southwest portion of Namibia, especially
the savanna region bordering Botswana and South Africa. Moreover, ignorance scores
could provide a simple way to quantify and visualise the impact of new expeditions on
biodiversity knowledge, providing a useful tool for demonstrating the value of such
enterprises. Indeed, it would be extremely interesting to annually re-evaluate ignorance
scores to provide a continuously updated account of progress in biological surveying
and data mobilization.
As demonstrated here and elsewhere (Ruete, 2015; Correia et al., 2019),
ignorance scores are also highly sensitive to spatial biases, making them useful tools to
identify socio-geographical factors influencing recording effort. Nevertheless, the
ignorance score algorithm also has certain limitations, the most serious of which is that
it could be considered overly simplistic for many forms of analysis since they are only
calculated using the number of records available in a given region over the time period
of analysis. This means that valuable information on, for example, the identities or
characteristics (e.g. threat status) of the species recorded, or the distribution of records
within the annual cycle are not considered (Meyer et al., 2016).
A particular limitation of the Log-Normalization algorithm is that the minimum
ignorance score (i.e. 0), is relative to the maximum number of records for the reference
taxonomic group (Ruete, 2015), which may still be low (for example, in our database
the maximum number of records in a cell for amphibians is 22). So, rather than
indicating complete biodiversity knowledge, an ignorance score of 0 should be
interpreted as the "best available knowledge" in any cell for the study region. In the
current study we attempted to counter the inherent simplicity of the algorithm by
independently considering multiple taxa and by restricting our analysis to more recent
records whose collection is likely to have been driven by similar socio-geographical
factors. In this context, ignorance scores provide a robust metric for measuring the
importance of data mobilization efforts on biodiversity knowledge, and we would
strongly recommend their use to quantify and visualize the impact of such initiatives.
Finally, it is important to highlight that the publicly available records on GBIF
for Namibia only represent a fraction of the documented biodiversity in the country.
Other types of institutions, such as museums, herbaria and other research centres in
Namibia and elsewhere harbour biodiversity data. For example, about 1.2 million bird
records were assembled during the first southern African bird atlas project (SABAP1)
95

which was pre-2000 and are not currently included in GBIF (J. Mendelsohn, pers.
comm.). In addition to the SABAP1 and SABAP2 projects (the latter already included
in GBIF), which have numerous records of birds, there are many other African projects
that use the efforts of citizen scientists to carry out inventories of taxonomic groups of
vertebrates, invertebrates, plants and even fungi, as is the case of the Virtual Museum
(Biodiversity and Development Institute, 2023).
The impediments to accessing, collecting and evaluating African biodiversity
data have been previously acknowledged and reported by decision-makers represented
by government, civil society and UN agencies, who have recommended a strengthening
of national and international collaboration to ensure availability and usability of
information and achieve conservation goals (Han et al, 2014; Stephenson et al, 2017).
Notwithstanding the significance of such engagement to collect and document data, it is
also imperative that these records be incorporated into online platforms and thus made
available for both scientists and decision makers.
Acknowledgements
TL was funded by the “Coordenação de Aperfeiçoamento de Pessoal de Nível Superior”
- Brazil (CAPES) - Finance Code 001. RAC acknowledges funding from the Kone
Foundation (#202101976) and the Academy of Finland (#348352). FA-M, JM-A,
ACMM and RJL is supported by the European Union’s Horizon 2020 research and
innovation programme under grant agreement no. 854248.
References
Atlas of Namibia Team. (2022). Atlas of Namibia: Its land, water and life. Namibia Nature
Foundation.
Ballesteros-Mejia, L., Kitching, I. J., Jetz, W., Nagel, P., & Beck, J. (2013). Mapping the
biodiversity of tropical insects: species richness and inventory completeness of African sphingid
moths. Global Ecology and Biogeography, 22(5), 586–595. https://doi.org/10.1111/geb.12039
Barve, V., & Otegui, J. (2016). bdvis: visualizing biodiversity data in R. Bioinformatics, 32(19),
3049–3050. https://doi.org/10.1093/bioinformatics/btw333
Biodiversity and Development Institute. (2023). The Virtual Museum. https://vmus.adu.org.za
Birch, C. P. D., Oom, S. P., & Beecham, J. A. (2007). Rectangular and hexagonal grids used for
observation, experiment and simulation in ecology. Ecological Modelling, 206(3–4), 347–359.
https://doi.org/10.1016/j.ecolmodel.2007.03.041

96

Blackie, R., Baldauf, C., Gautier, D., Gumbo, D., Kassa, H., Parthasarathy, N., Paumgarten, F.,
Sola, P., Pulla, S., Waeber, P., & Sunderland, T. (2014). Tropical dry forests: The state of global
knowledge and recommendations. Center for International Forestry Research.
Boakes, E. H., Mcgowan, P. J. K., Fuller, R. A., Chang-qing, D., Clark, N. E., Connor, K. O., &
Mace, G. M. (2010). Distorted views of biodiversity: spatial and temporal bias in species
occurrence data. PLoS Biology, 8(6). https://doi.org/10.1371/journal.pbio.1000385
Bonney, R. (2021). Expanding the Impact of Citizen Science. BioScience, 71(5), 448–451.
https://doi.org/10.1093/biosci/biab041
Carvalho, R.L., Resende, A.F., Barlow, J., et al. (2023) Pervasive gaps in Amazonian ecological
research. Current Biology, S0960982223008631.
Chamberlain, S. A., & Boettiger, C. (2017). R Python, and Ruby clients for GBIF species
occurrence
data.
PeerJ
Preprint,
5,
e3304v1.
https://doi.org/https://doi.org/10.7287/peerj.preprints.3304v1
Chamberlain, S. A., Szoecs, E., Foster, Z., Arendsee, Z., Boettiger, C., Ram, K., Bartomeus, I.,
Baumgartner, J., O’Donnell, J., Oksanen, J., Tzovaras, B. G., Marchand, P., Tran, V., Salmon,
M., Li, G., & Grenié, M. (2020). taxize: Taxonomic information from around the web (R
package version 0.9.98). https://github.com/ropensci/taxize
Chandler, M., See, L., Copas, K., Bonde, A. M. Z., López, B. C., Danielsen, F., Legind, J. K.,
Masinde, S., Miller-Rushing, A. J., Newman, G., Rosemartin, A., & Turak, E. (2017).
Contribution of citizen science towards international biodiversity monitoring. Biological
Conservation, 213, 280–294. https://doi.org/10.1016/j.biocon.2016.09.004
Cobos, M. E., Jiménez, L., Nuñez-Penichet, C., Romero-Alvarez, D., & Simoes, M. (2018).
Sample data and training modules for cleaning biodiversity information. Biodiversity
Informatics, 13, 49–50. https://doi.org/10.17161/bi.v13i0.7600
Correia, R. A., Ruete, A., Stropp, J., Malhado, A. C. M., dos Santos, J. W., Lessa, T., Alves, J.
A., & Ladle, R. J. (2019). Using ignorance scores to explore biodiversity recording effort for
multiple taxa in the Caatinga. Ecological Indicators, 106(June 2019), 105539.
https://doi.org/10.1016/j.ecolind.2019.105539
Corrigan, C., Bingham, H., Shi, Y., Lewis, E., Chauvenet, A., & Kingston, N. (2018).
Quantifying the contribution to biodiversity conservation of protected areas governed by
indigenous peoples and local communities. Biological Conservation, 227, 403–412.
https://doi.org/10.1016/j.biocon.2018.09.007
D’Antraccoli, M., Bedini, G., & Peruzzi, L. (2022). Maps of relative floristic ignorance and
virtual floristic lists: An R package to incorporate uncertainty in mapping and analysing
biodiversity
data.
Ecological
Informatics,
67,
101512.
https://doi.org/10.1016/j.ecoinf.2021.101512
Danovaro, R., Company, J. B., Corinaldesi, C., D’Onghia, G., Galil, B., Gambi, C., Gooday, A.
J., Lampadariou, N., Luna, G. M., & Morigi, C. (2010). Deep-sea biodiversity in the
Mediterranean Sea: the known, the unknown, and the unknowable. PloS One, 5(8), e11832.
dos Santos, J. G., Malhado, A. C. M., Ladle, R. J., Correia, R. A., & Costa, M. H. (2015).
Geographic trends and information deficits in Amazonian conservation research. Biodiversity
and Conservation, 24(11), 2853–2863. https://doi.org/10.1007/s10531-015-0981-x

97

Edwards, J. L. (2004). Research and Societal Benefits of the Global Biodiversity Information
Facility. BioScience, 54(6), 485–486. https://doi.org/, https://doi.org/10.1641/0006-3568
Escribano, N., Ariño, A.H. and Galicia, D. (2016). Biodiversity data obsolescence and land uses
changes. PeerJ, 4, p.e2743. doi:https://doi.org/10.7717/peerj.2743.
Gaiji S., Chavan, V., Ariño, A. H., Otegui, J., Hobern, D., Sood, R., & Robles, E. (2013).
Content assessment of the primary biodiversity data published through GBIF network: Status,
challenges and potentials. Biodiversity Informatics, 8(2). https://doi.org/10.17161/bi.v8i2.4124
Gargallo, E. (2020). Community Conservation and Land Use in Namibia: Visions, Expectations
and
Realities.
Journal
of
Southern
African
Studies,
46(1),
129–147.
https://doi.org/10.1080/03057070.2020.1705617
GBIF.
(2021).
Namibia
https://doi.org/https://doi.org/10.15468/dl.9uwfx9

Occurrence

Download.

GBIF. (2023). GBIF Home Page. https://www.gbif.org/grscicoll/institution/search
Giess, W. (1971). A preliminary vegetation map of South West Africa. Dinteria, 4, 5–14.
Gotelli, N. J., Booher, D. B., Urban, M. C., Ulrich, W., Suarez, A. V., Skelly, D. K., Russell, D.
J., Rowe, R. J., Rothendler, M., Rios, N., Rehan, S. M., Ni, G., Moreau, C. S., Magurran, A. E.,
Jones, F. A. M., Graves, G. R., Fiera, C., Burkhardt, U., & Primack, R. B. (2021). Estimating
species relative abundances from museum records. Methods in Ecology and Evolution, 14(2),
431–443. https://doi.org/10.1111/2041-210X.13705
Han, X., Smyth, R. L., Young, B. E., Brooks, T. M., Sánchez de Lozada, A., Bubb, P., Butchart,
S. H. M., Larsen, F. W., Hamilton, H., Hansen, M. C., & Turner, W. R. (2014). A Biodiversity
Indicators Dashboard: Addressing Challenges to Monitoring Progress towards the Aichi
Biodiversity Targets Using Disaggregated Global Data. PLoS ONE, 9(11), e112046.
https://doi.org/10.1371/journal.pone.0112046
Hansen, M. C., Potapov, P. V., Moore, R., Hancher, M., Turubanova, S. A., Tyukavina, A.,
Thau, D., Stehman, S. V., Goetz, S. J., Loveland, T. R., Kommareddy, A., Egorov, A., Chini,
L., Justice, C. O., & Townshend, J. R. G. (2013). High-Resolution Global Maps of 21st-Century
Forest Cover Change. Science, 342(6160), 850–853. https://doi.org/10.1126/science.1244693
Hedrick, B. P., Heberling, J. M., Meineke, E. K., Turner, K. G., Grassa, C. J., Park, D. S.,
Kennedy, J., Clarke, J. A., Cook, J. A., Blackburn, D. C., Edwards, S. V, & Davis, C. C. (2020).
Digitization and the Future of Natural History Collections. BioScience, 70(3), 243–251.
https://doi.org/10.1093/biosci/biz163
Hockings, M., Dudley, N., & Elliott, W. (2020). COVID-19 and protected and conserved areas.
PARKS, 26.1, 7–24. https://doi.org/10.2305/IUCN.CH.2020.PARKS-26-1MH.en
Hopkins, M. J. G. (2019). Are we close to knowing the plant diversity of the Amazon? Anais
Da Academia Brasileira de Ciências, 91.
Hortal, Joaquín, Bello, F. de, Diniz-Filho, J. A. F., Lewinsohn, T. M., Lobo, J. M., & Ladle, R.
J. (2015). Seven shortfalls that beset large-scale knowledge of biodiversity. Annu. Rev. Ecol.
Evol. Syst., 46, 523–552. https://doi.org/10.1146/annurev-ecolsys-112414-054400
Hortal, Joaquín, Ladle, R. J., Stropp, J., & Tessarolo, G. (2022). Accounting for biogeographical
ignorance within biodiversity modelling. Research Outreach, 129. https://doi.org/10.32907/RO129-2502594871
98

Kadmon, R., Farber, O., & Danin, A. (2004). Effect of roadside bias on the accuracy of
predictive maps produced by bioclimatic models. Ecological Applications, 14(2), 401–413.
Klintenberg, P., & Seely, M. (2004). Land Degradation Monitoring in Namibia: A First
Approximation.
Environmental
Monitoring
and
Assessment,
99(1–3),
5–21.
https://doi.org/10.1007/s10661-004-3994-6
Ladle, R., & Hortal, J. (2013). Mapping species distributions: living with uncertainty. Frontiers
of Biogeography, 5(1). https://doi.org/10.21425/F55112942
Lessa, T., Dos Santos, J. W., Correia, R. A., Ladle, R. J., & Malhado, A. C. M. (2019). Known
unknowns: Filling the gaps in scientific knowledge production in the Caatinga. PLoS ONE,
14(7), 1–12. https://doi.org/10.1371/journal.pone.0219359
Liu, J., Zhao, Y., Si, X., Feng, G., Slik, F., & Zhang, J. (2021). University campuses as valuable
resources for urban biodiversity research and conservation. Urban Forestry & Urban Greening,
64, 127255. https://doi.org/10.1016/j.ufug.2021.127255
Luck, G. W. (2007). A review of the relationships between human population density and
biodiversity. Biological Reviews, 82(4), 607–645. https://doi.org/10.1111/j.1469185X.2007.00028.x
Mair, L., & Ruete, A. (2016). Explaining Spatial Variation in the Recording Effort of Citizen
Science
Data
across
Multiple
Taxa.
PLOS
ONE,
11(1),
e0147796.
https://doi.org/10.1371/journal.pone.0147796
Meyer, C., Kreft, H., Guralnick, R., & Jetz, W. (2015). Global priorities for an effective
information basis of biodiversity distributions. Nature Communications, 6(1), 8221.
https://doi.org/10.1038/ncomms9221
Meyer, C. (2016). Limitations in global information on species occurrences. Frontiers of
Biogeography, 8(2), e28195.
Meyer, C., Weigelt, P., & Kreft, H. (2016). Multidimensional biases, gaps and uncertainties in
global
plant
occurrence
information.
Ecology
Letters,
19(8),
992–1006.
https://doi.org/10.1111/ele.12624
Millar, E. E., Hazell, E. C., & Melles, S. J. (2019). The ‘cottage effect’ in citizen science?
Spatial bias in aquatic monitoring programs. International Journal of Geographical Information
Science, 33(8), 1612–1632. https://doi.org/10.1080/13658816.2018.1423686
NSA (2023). Namibia Statistics Agency. https://nsa.nsa.org.na/
Nelson, G., & Ellis, S. (2019). The history and impact of digitization and digital data
mobilization on biodiversity research. Philosophical Transactions of the Royal Society B:
Biological Sciences, 374(1763), 20170391. https://doi.org/10.1098/rstb.2017.0391
Oliveira, U., Paglia, A. P., Brescovit, A. D., Carvalho, C. J. B. De, Silva, D. P., Rezende, D. T.,
Stehmann, R., Pena, P., & Remoto, C. D. S. (2016). The strong influence of collection bias on
biodiversity knowledge shortfalls of Brazilian terrestrial biodiversity. Diversity and
Distributions, 1–13. https://doi.org/10.1111/ddi.12489
Ospina, R., & Ferrari, S. L. P. (2010). Inflated beta distributions. Statistical Papers, 51(1), 111–
126. https://doi.org/10.1007/s00362-008-0125-4

99

Petersen, T. K., Speed, J. D. M., Grøtan, V., & Austrheim, G. (2021). Species data for
understanding biodiversity dynamics: The what, where and when of species occurrence data
collection. Ecological Solutions and Evidence, 2(1). https://doi.org/10.1002/2688-8319.12048
Phillips, S. J., Dudík, M., Elith, J., Graham, C. H., Lehmann, A., Leathwick, J., & Ferrier, S.
(2009). Sample selection bias and presence-only distribution models: implications for
background and pseudo-absence data. Ecological Applications, 19(1), 181–197.
https://doi.org/10.1890/07-2153.1
Ponder, W. F., Carter, G. A., Flemons, P., & Chapman, R. R. (2001). Evaluation of Museum
Collection Data for Use in Biodiversity Assessment. Conservation Biology, 15(3), 648–657.
https://doi.org/10.1046/j.1523-1739.2001.015003648.x
R Team Core. (2017). R: A Language and Environment for Statistical Computing. R
Foundation for Statistical Computing.
Reddy, S., & Dávalos, L. M. (2003). Geographical sampling bias and its implications for
conservation priorities in Africa. Journal of Biogeography, 30(11), 1719–1727.
https://doi.org/10.1046/j.1365-2699.2003.00946
Revermann, R., Wallenfang, J., Oldeland, J., & Finckh, M. (2017). Species richness and
evenness respond to diverging land-use patterns - a cross-border study of dry tropical
woodlands in southern Africa. African Journal of Ecology, 55(2), 152–161.
https://doi.org/10.1111/aje.12333
Ribeiro, G.V.T., Teixido, A.L., Barbosa, N.P.U. & Silveira, F.A.O. (2016) Assessing bias and
knowledge gaps on seed ecology research: implications for conservation agenda and policy.
Ecological Applications, 26, 2033–2043.
Rigby, R. A., & Stasinopoulos, D. M. (2005). Generalized additive models for location, scale
and shape (with discussion). Journal of the Royal Statistical Society: Series C (Applied
Statistics), 54(3), 507–554. https://doi.org/10.1111/j.1467-9876.2005.00510.x
Rocchini, D., Hortal, J., Lengyel, S., Lobo, J. M., Jiménez-Valverde, A., Ricotta, C., Bacaro, G.,
& Chiarucci, A. (2011). Accounting for uncertainty when mapping species distributions: The
need for maps of ignorance. Progress in Physical Geography, 35(2), 211–226.
https://doi.org/10.1177/0309133311399491
Rocha‐Ortega, M., Rodriguez, P., & Córdoba‐Aguilar, A. (2021). Geographical, temporal and
taxonomic biases in insect GBIF data on biodiversity and extinction. Ecological Entomology,
46(4), 718–728. https://doi.org/10.1111/een.13027
Ruete, A. (2015). Displaying bias in sampling effort of data accessed from biodiversity
databases
using
ignorance
maps.
Biodiversity
Data
Journal,
3(1).
https://doi.org/10.3897/BDJ.3.e5361
Sastre, P., & Lobo, J. M. (2009). Taxonomist survey biases and the unveiling of biodiversity
patterns.
Biological
Conservation,
142(2),
462–467.
https://doi.org/10.1016/j.biocon.2008.11.002
Schalkwyk, D. L. van, McMillin, K. W., Witthuhn, R. C., & Hoffman, L. C. (2010). The
contribution of wildlife to sustainable natural resource utilization in Namibia: a review.
Sustainability, 2(11), 3479–3499. https://doi.org/10.3390/su2113479

100

Simmons, R. E., Griffin, M., Griffin, R. E., Marais, E., & Kolberg, H. (1998). Endemism in
Namibia: patterns, processes and predictions. Biodiversity and Conservation, 7(513–530).
https://doi.org/https://doi.org/10.1023/A:1008879712736
Sousa-Baena, M. S., Garcia, L. C., & Peterson, A. T. (2014b). Completeness of digital
accessible knowledge of the plants of Brazil and priorities for survey and inventory. Diversity
and Distributions, 20(4), 369–381. https://doi.org/10.1111/ddi.12136
Souza, C. N., Rodrigues, A. C., Correia, R. A., Normande, I. C., Costa, H. C. M., GuedesSantos, J., Malhado, A. C. M., Carvalho, A. R., & Ladle, R. J. (2021). No visit, no interest: How
COVID-19 has affected public interest in world’s national parks. Biological Conservation, 256,
109015. https://doi.org/10.1016/j.biocon.2021.109015
Stasinopoulos, D. M., & Rigby, R. A. (2007). Generalized Additive Models for Location Scale
and
Shape
(GAMLSS)
in
R.
Journal
of
Statistical
Software,
23(7).
https://doi.org/10.18637/jss.v023.i07
Stasinopoulos, D. M., Rigby, R. A., Heller, G. Z., Voudouris, V., & Bastiani, F. De. (2017).
Flexible Regression and Smoothing: Using GAMLSS in R (1st ed.). Chapman and Hall/CRC.
https://doi.org/10.1201/b21973
Steege, H. t., Haripersaud, P. P., Banki, O. S., & Schieving, F. (2011). A model of botanical
collectors’ behavior in the field: Never the same species twice. American Journal of Botany,
98(1), 31–37. https://doi.org/10.3732/ajb.1000215
Stephenson, P. J., Bowles-Newark, N., Regan, E., Stanwell-Smith, D., Diagana, M., Höft, R.,
Abarchi, H., Abrahamse, T., Akello, C., Allison, H., Banki, O., Batieno, B., Dieme, S.,
Domingos, A., Galt, R., Githaiga, C. W., Guindo, A. B., Hafashimana, D. L. N., Hirsch, T., &
Hobern, D. (2017). Unblocking the flow of biodiversity data for decision-making in Africa.
Biological Conservation, 213, 335–340. https://doi.org/10.1016/j.biocon.2016.09.003
Stropp, J., Ladle, R. J., M. Malhado, A. C., Hortal, J., Gaffuri, J., H. Temperley, W., Olav
Skøien, J., & Mayaux, P. (2016a). Mapping ignorance: 300 years of collecting flowering plants
in
Africa.
Global
Ecology
and
Biogeography,
25(9),
1085–1096.
https://doi.org/10.1111/geb.12468
Tessarolo, G., Ladle, R., Rangel, T., & Hortal, J. (2017). Temporal degradation of data limits
biodiversity research. Ecology and Evolution, 7(17), 6863–6870.
Tessarolo, G., Ladle, R., Lobo, J. M., Rangel, T. F., & Hortal, J. (2021). Using maps of
biogeographical ignorance to reveal the uncertainty in distributional data hidden in species
distribution models. Ecography, 44(12), 1743–1755. https://doi.org/10.1111/ecog.05793
The
World
Bank.
(2022).
Population,
total
https://data.worldbank.org/indicator/SP.POP.TOTL?locations=NA

-

Namibia.

Wardell-Johnson, G. (2000). Biodiversity and Conservation in Namibia into the 21st Century.
In Population–Development–Environment in Namibia (pp. 17–47). International Institute for
Applied Systems Analysis.
Wart, M. Van, Hondeghem, A., Schwella, E., & Suino, P. (2015). Leadership and Culture:
Comparative Models of Top Civil Servant Training (M. Van Wart, A. Hondeghem, E. Schwella,
& P. Suino (eds.)). Palgrave Macmillan UK. https://doi.org/10.1057/9781137454133

101

Yang, W., Ma, K., & Kreft, H. (2014). Environmental and socio-economic factors shaping the
geography of floristic collections in China. Global Ecology and Biogeography, 23(11), 1284–
1292. https://doi.org/10.1111/geb.12225
Zizka, A., Silvestro, D., Andermann, T., Azevedo, J., Duarte Ritter, C., Edler, D., Farooq, H.,
Herdean, A., Ariza, M., Scharn, R., Svantesson, S., Wengström, N., Zizka, V., & Antonelli, A.
(2019). CoordinateCleaner: Standardized cleaning of occurrence records from biological
collection databases. Methods in Ecology and Evolution, 10(5), 744–751.
https://doi.org/10.1111/2041-210X.13152
Zizka, A., Antunes Carvalho, F., Calvente, A., Rocio Baez-Lizarazo, M., Cabral, A., Coelho,
J.F.R., Colli-Silva, M., Fantinati, M.R., Fernandes, M.F., Ferreira-Araújo, T., Gondim Lambert
Moreira, F., Santos, N.M.C., Santos, T.A.B., Dos Santos-Costa, R.C., Serrano, F.C., Alves Da
Silva, A.P., De Souza Soares, A., Cavalcante De Souza, P.G., Calisto Tomaz, E., Vale, V.F.,
Vieira, T.L. & Antonelli, A. (2020) No one-size-fits-all solution to clean GBIF. PeerJ, 8, e9916.

102

Conclusões gerais
Esta tese contribui para expandir a discussão sobre os principais motivadores de
ignorância e incerteza nos dados taxonômicos e espaciais da biodiversidade. No
primeiro capítulo, apresentamos uma perspectiva sobre como as mudanças taxonômicas,
ocasionada pelo progresso nas práticas taxonômicas, têm levado a flutuações e
imprecisões sobre o número de espécies já reconhecidas e o número espécies que
realmente existente no mundo (lacuna Linneana). As revisões taxonômicas têm
colaborado para os processos de divisões e agrupamentos taxonômicos, que por sua vez
afetam as estimativas de riqueza de espécies. Nós recomendamos que cientistas sejam
cautelosos na condução e interpretação das estimativas globais, mas também que seja
imperativa a inclusão de dados sobre as mudanças taxonômicas nas estimativas da
biodiversidade atual. Enfatizamos que a lacuna Linneana é um fenômeno biocultural,
que será afetada por diversos aspectos como a história da área explorada e o esforço
despendido na busca por táxons, ou ainda, como as espécies são definidas e delimitadas
ou o nível de digitalização/mobilização de dados.
No segundo capítulo, segundo nossa métrica para avaliar da incerteza
nomenclatural, observamos que mais de metade (56%) das espécies de aves globais
apresentaram algum grau de incerteza relacionada ao nome científico. O cenário foi
mais crítico nas classificações taxonômicas superiores, pois apenas seis das 36 Ordens
de aves avaliadas não apresentaram incerteza nomenclatural. As variáveis biológicas e
ecológicas que foram associadas com a incerteza nomenclatural foram: massa corporal,
tamanho da área de distribuição, distinção evolutiva e status de conservação da IUCN.
Sendo as espécies que apresentaram menores porcentagens de incerteza taxonômica
àquelas de maior tamanho corporal, ampla distribuição geográfica, espécies não
ameaçadas de extinção e espécies mais distintas evolutivamente. Discutimos que a
taxonomia é uma ciência dinâmica e que as discordâncias de nomenclatura são reflexos
de progresso, porém, também da falta de consenso da comunidade científica sobre o
estatuto das unidades taxonômicas.
No terceiro capítulo, observamos uma desigualdade taxonômica e espacial nos
registros de ocorrências da Namíbia, África. A maior parte (94%) dos registros de
ocorrência dos cinco grupos taxonômicos de referência disponíveis no GBIF foi de
Aves, especialmente devido ao influxo de dados em plataforma de ciência cidadã
(eBird). Além disso, para maioria dos grupos taxonômicos avaliados, muitos dados
103

podem ser considerados obsoletos, pois foram coletados antes dos anos 2000.
Observamos tendências espaciais nas pontuações de ignorância que foram
particularmente influenciadas pela maior densidade de rodovias, proximidade de
instituições de pesquisa e de áreas protegidas. Destacamos a importância da ciência
cidadã na mobilização de dados da biodiversidade, mas observamos que ainda são
necessários esforços conjuntos de instituições (museus, herbários e outros centros de
investigação) para acessar e disponibilizar dados da biodiversidade.
Nossos resultados, especialmente nos dos dois últimos capítulos, demonstram
que muitos cientistas e coletadores estão realizando investigações taxonômicas, seja em
suas zonas de conveniência e afinidade, mas principalmente pela falta de investimentos
para expedições de campo, estudos taxonômicos, desenvolvimento e capacitação em
novas tecnologias. Além disso, há impedimentos no acesso, recolha e usabilidade dos
dados da biodiversidade, sendo extremamente recomendado que cientistas e instituições
tornem os dados disponíveis para toda comunidade científica e tomadores de decisão.
Finalmente, esta tese reforça a importância de investir em ciência de base para garantir
que o conhecimento taxonômico e espacial das espécies seja amplamente difundido e as
lacunas preenchidas.

104