Here we worked on the rails (bird family Rallidae), looking at whether the current threats are consistent with those that led to recent extinctions, and ultimately, what conservation actions might be necessary to mitigate further losses. We undertook a global synthesis of the temporal and spatial threat patterns for Rallidae and determined conservation priorities and gaps.

Database compilation

We compiled a database for all 144 species of extant rails (including 42 island endemic species and 33 threatened species) and 24 extinct rails using the 2019 version of the International Union for Conservation of Nature (IUCN) Red List (IUCN, 2019) and the ‘Guide to the Rails, Crakes, Gallinules and Coots of the world’ (Taylor & van Perlo, 1998). We used the taxonomic classification followed by the IUCN which included the rallid family of Sarothruridae (some authors consider it separate from the Rallidae family, see (Garcia-R. et al., 2014; Garcia-R et al., 2020)). The Sarothruridae family contains 15 species, including two threatened species, and is mostly present in the Afrotropics. In preliminary analyses, we evaluated the effect of excluding this group, finding no major effect on results or their interpretations (except a minor effect of Threat rank of countries, see Table S1) and, thus, elected to retain the Sarothruridae in the Rallidae family. We considered island endemic species as those restricted to one or a group of adjacent islands. Species considered ‘Data deficient’ were excluded from the analysis (Table S2). Similarly, the New Caledonian rail (Gallirallus lafresnayanus) and the Samoan moorhen (Pareudiastes pacificus) are two ‘Critically endangered’ rail species that have not been seen with certainty since the 19^th century and are suspected to be extinct (IUCN, 2019), so they were considered ‘extinct’ in this study (Table S2).

Spatial and temporal patterns of threats

We compiled the contemporary threats for all extant rails and historical causes of extinction for all extinct rails, available from the online IUCN database for each species (http://iucnredlist.org). The IUCN’s threats classification organises threats in a hierarchy (https://www.iucnredlist.org/resources/threat-classification-scheme).

2.1 Spatial threats patterns of extinct and threatened rails

First, we compared the historical causes of extinction (extinct rails) to the contemporary threats (threatened rails) to assess temporal changes in the threat pattern and changes in the spatial distribution of threatened and extinct species.

Then, we described the number of threats impacting threatened and extinct species for different spatial scales (globally, on continents, on islands, between bioregions). For this analysis of threat diversity, we used the threats’ first sub-category as the level of threat (e.g., a rail being impacted by sub-categories ‘5.1. Hunting & collecting terrestrial animals’ and ‘5.3. Logging & wood harvesting’ within the threat ‘5. Biological resource use’ was considered as threatened by two threats). A total of 45 threats were considered, but only 28 were relevant to rails.

2.2 Spatial threats patterns of all extant rails

We assessed the spatial pattern of contemporary threats for all extant rails (whether or not they were considered threatened by the IUCN) by descriptively comparing the type of threats and their impact: i) globally, ii) on islands, iii) on continents, and iv) by bioregions (biogeographic realms, hereafter ‘bioregions’: Australasia, Oceania, Nearctic, Neotropics, Palearctic, Afrotropics, Indomalaya; defined following (Olson et al., 2001)). Oceania was grouped with the Australasia bioregion because of its few extant species (five, including three present across the two bioregions).

We calculated each threat’s impact score using the Threat Impact Scoring System (IUCN - CMP Unified Classification of Direct Threats, version 3.2) proposed by the IUCN (2019) that defines an overall impact based on scope, severity and timing, and ranges within ‘Negligible’, ‘Low’, ‘Medium’, and ‘High’ impact (https://www.iucnredlist.org/resources/threat-classification-scheme; Fig. S1). If a threat had a different impact in two sub-categories, we used the higher impact for the classification. We included ‘Past’ and ‘Ongoing’ impact to illustrate the temporal evolution of threats (‘Future’ was not considered).

For this analysis, we used the first category as the level of threat (e.g., 1. ‘Residential & commercial development’ and 2. ‘Agriculture’) to be more informative. We split the threat ‘5. Biological resource use’ in two categories: ‘Hunting & collecting terrestrial animals’ (5.1.) and ‘Logging & indirect effects’ (regrouping ‘5.2. Gathering terrestrial plants’, ‘5.3 Logging & wood harvesting’, and ‘5.4. Fishing & harvesting aquatic resources’) to provide relevant conservation policy. The sub-category ‘10. Geological events’ was not included because it was not listed for any species.

Conservation status and gaps

To identify gaps in conservation efforts for rails, we summarised the IUCN’s Conservation actions classification scheme version 2.0 (https://www.iucnredlist.org/resources/conservation-actions-classification-scheme) and Research needed classification scheme version 2.0 (https://www.iucnredlist.org/resources/research-needed-classification-scheme): i) globally, ii) per country, and iii) per bioregion, following Olah et al. (2016). Countries and overseas territories were grouped together for country-level analysis (Table S3). These schemes specify the conservation actions and research needed for each species and were available on the online IUCN database (http://iucnredlist.org). We gathered the possible classifications in five categories: (i) ‘Research & Monitoring’, (ii) ‘Ecosystem Protection & Management’, (iii) ‘Species Management’, (iv) ‘Education & Awareness’, and (v) ‘Law & Policy’ (i.e., legislative protection).

(i) ‘Research & Monitoring’ category was formed by grouping all categories in the Research Needed Classification Scheme: ‘1. Research’, ‘2. Conservation Planning’, ‘3. Monitoring’, and ‘4. Other’. (ii) ‘Ecosystem Protection & Management’ category was formed by grouping the first two categories, ‘1. Land/water protection’ and ‘2. Land/water management’, in the Conservation Actions Classification Scheme. (iii) ‘Species Management’, (iv) ‘Education & Awareness’, and (v) ‘Law & Policy’ corresponded directly to each remaining category in the Conservation Actions Classification Scheme. Furthermore, category ‘6. Livelihood, economic & other incentives’ of the Conservation Actions Classification Scheme was not represented as it did not appear in any of the studied species.

To assess whether conservation status had improved or deteriorated over time, we characterised changes in species’ IUCN conservation status since 1988 (first IUCN Red List assessment) through to 2019 (IUCN, 2019). Changes in conservation status could happen any year between 1988 and 2019. In cases where status changed more than once, we only used the most recent change. We considered ‘Unknown (LR/LC)’ as Least Concern and ‘Unknown (LR/NT)’ as Near Threatened when they were the assessments preceding a new status. However, we ignored it when included between two assessments. For example, if a species conservation status’ history was: 1996: Vulnerable (VU); 2000: Unknown (LR/NT); 2004: Near threatened (NT), we considered that the species improved from Vulnerable to Near threatened.

Rails ‘conservation hotspots’ and priority rankings

Rallidae combine species with unique evolutionary traits (e.g., flightlessness and endemism) and species with elevated vulnerability (24% of the species are threatened); however to date, there is no framework that accounts for these important aspects in terms of conservation priority. To identify areas with high conservation interest and/or those that deserve improvements in their protection, we conceived a ranking system to classify both world bioregions and countries of high conservation priority for rails (‘conservation hotspots’). Analysis at the country-level grouped countries and their overseas territories together (Table S3). We created two categories that we deemed useful to reflect important rail conservation: ‘Heritage’ and ‘Threat’.

The ‘Heritage’ category aimed to account for species with high conservation value, using endemism and unique evolutionary trait (flightlessness). It was calculated by ranking the number of rail species being either flightless (firstly), island endemic (secondly), or country endemic (thirdly). When relevant, species were classified by one of the above attributes, in this hierarchical order. For instance, a flightless rail that is also endemic to an island, would only be classified as ‘flightless’, and an island endemic rail that is also endemic to one country, would only be classified as ‘island endemic’.

The ‘Threat’ category was calculated by ranking the number of threatened species (firstly), and the number of species with a worsened IUCN status since their last change in status (secondly).

In cases where ranks were equal, they were split using the country/bioregion’s richness in rail species as the tie breaker (i.e., higher richness would get the higher rank of conservation importance). In this rank system, rank one represents the highest rank for conservation priority. The rank classifications were not made to adequately differentiate between two entities as close scores would not illustrate true difference in conservation priority. We recommend identifying entities with the highest conservation priority as the ones present in both top classifications for the two ranks, or as parts of a top 5, top 10, or top 20.

REFERENCES

Garcia-R, J.C., Lemmon, E.M., Lemmon, A.R. & French, N. (2020) Phylogenomic Reconstruction Sheds Light on New Relationships and Timescale of Rails (Aves: Rallidae) Evolution. Diversity, 12, 70. https://doi.org/10.3390/d12020070

Garcia-R., J.C., Gibb, G.C. & Trewick, S.A. (2014) Deep global evolutionary radiation in birds: diversification and trait evolution in the cosmopolitan bird family Rallidae. Molecular Phylogenetics and Evolution, 81, 96-108. https://doi.org/10.1016/j.ympev.2014.09.008

IUCN (2019) The IUCN Red List of Threatened Species. Version 2019-3. Available at:  http://www.iucnredlist.org (accessed 12 December 2019)

Olah, G., Butchart, S.H.M., Symes, A., Guzmán, I.M., Cunningham, R., Brightsmith, D.J. & Heinsohn, R. (2016) Ecological and socio-economic factors affecting extinction risk in parrots. Biodiversity and Conservation, 25, 205-223. https://doi.org/10.1007/s10531-015-1036-z

Olson, D.M., Dinerstein, E., Wikramanayake, E.D., Burgess, N.D., Powell, G.V., Underwood, E.C., D'amico, J.A., Itoua, I., Strand, H.E. & Morrison, J.C. (2001) Terrestrial Ecoregions of the World: A New Map of Life on Earth. A new global map of terrestrial ecoregions provides an innovative tool for conserving biodiversity. BioScience, 51, 933-938. https://doi.org/10.1641/0006-3568(2001)051[0933:TEOTWA]2.0.CO;2

Taylor, B. & van Perlo, B. (1998) Rails: a guide to rails, crakes, gallinules and coots of the world. Pica Press, Mountfield, U.K.

Date of data collection: 2017-2020

Date file was written: 2021/02/01

For more information on methods, please refer to publication preprint (https://doi.org/10.1101/2021.01.20.427508).

All modelling is done in R software.

The dataset is separated into two parts:

1. THREAT PATTERN: look at characterising how threats are distributed at different scales (bioregions, global and islands), the threat diversity per species, and compare their intensity spatially. We also looked at the causes of extinctions in extinct rails from the 16th century onwards to compare with modern threat pattern.

Variables in code are:

1.1. THREAT DIVERSITY (R.file 'Leveque et al. threat diversity')

-iucn: iucn conservation status: T= threatened (include 'vulnerable', 'endangered', 'critically endangered'); NoT= not threatened ('not concerned', 'near threatened')

-common.name: vernacular name of the species

-island.ende: whether the species is island endemic (value: "yes" or "no")

-flightless: whether  the species is flightless (value: "flightless" or "flying")

-threatdiv: how many threats each species is suffering from

-scale: the spatial scale considered (value: 'gl.combined': global scale (both continental and island species, including threatened ('T') and non-threatened ('NoT') species; 'global': both continental and island species; 'continent': continental species only; 'island': island endemic species only)

-type: the spatial scale with the threatened status of a species (value: 'combined': threatened and non-threatened species together; 'globalT': global scale for threatened species only; 'globalNoT': global scale for non-threatened species only; 'continent-T': continental species that are threatened only; 'continent-NoT': continental species that are non-threatenend only; 'island-T': island endemic species that are threatened only; 'island-NoT': island endemic species that are non-threatened only)

1.2. THREAT PATTERNS (R.files 'Leveque et al. threats_global and island' and 'Leveque et al. threats_regions')

-threat: threats impacting species (value according to the ICUN: "Residential & commercial development"; "Agriculture"; "Energy production and mining"; "Transportation & service corridors"; "Hunting & collecting animals"; "Logging & wood harvesting"; "Human intrusions & disturbance"; "Natural system modifications"; "Invasive & problematic species"; "Pollution"; "Climate change & severe weather"; "Other")

-impact: the strength of the threat impact, according to the IUCN (value: "Low Impact", "Medium Impact", "Unknown", "Past Impact")

-percentage: percentage of species impacted by the threat at this scale

-nb. sp: the number of species impacted by the threat at this scale

-scale: scale considered (value: "Gb": global; "Is": island endemic species; "Ct": continental species)

REGIONS

-region: bioregion a species is found in (value: Afrotropic; Australasia/Oceania; Indomalaya; Nearctic; Neotropic ;Other; Palearctic. Other refers to remote islands in the Atlantic ocean that were hardly associable with a bioregion. (note: the same species can be present in different bioregions)

-iucn: iucn conservation status: 'VU' (vulnerable); EN (endangered); CR (critically endangered); LC (least concerned), NT (near threatened)

2. RANKINGS AND GAPS: Look at describing what conservation gaps are missing per rail species (based on the IUCN database, available on their website online).
Secondly, we created a ranking system to find what countries and what bioregions are in priority for conservation (the ones that need to organise conservation efforts the most). The ranking system is based on the degree of threat and threatening clues (having a degraded IUCN status; 'Threat rank'). The other ranking system is based on the uniqueness of species (flightless, island endemic or country endemic) for their cultural value ('Heritage' rank) that therefore should have inforced conservation efforts. The two ranks are analysed separately, but in the case of countries present in two ranks, these are of the highest priority for conservation actions.

R.files are: 'Leveque et al. 2021 - ranking_gaps_countries' and 'Leveque et al. 2021 - ranking_gaps_regions_2020')
Variables in R code are:

-iucn: iucn conservation status: T= threatened (include 'vulnerable', 'endangered', 'critically endangered'); NoT= not threatened ('least concerned', 'near threatened').

-country: country name where rail species are present (note: the same species can be present in different countries)

-n: number of countries

-bioregion: bioregion a species is found in (value: Afrotropic; Australasia/Oceania; Indomalaya; Nearctic; Neotropic ;Other; Palearctic. Other refers to remote islands in the Atlantic ocean that were hardly associable with a bioregion. (note: the same species can be present in different bioregions)

-research: gap in ‘Research, Monitoring and Planning’(value: 'needed' (later '1') and 'no (later '0'))

-habitat.prot: gap in ‘Ecosystem Protection and Management’(value: 'needed' (later '1') and 'no (later '0'))

-management: gap in ‘Species Management’ (value: 'needed' (later '1') and 'no (later '0'))

-education : gap in ‘Education & Awareness’ (value: 'needed' (later '1') and 'no (later '0'))

-law : gap in ‘Law & Policy’ (i.e., legislative protection) (value: 'needed' (later '1') and 'no (later '0'))

-lack: average number of conservation gaps per threatened sp.

-degraded.status: if the iucn conservation status has deteriorated (value: 'improved'  or 'yes')

-richness: countries sp richness (rails) (value: number of rails species)

-island.ende: whether the species is island endemic (value: "yes" or "no"

-flightless: whether  the species is flightless (value: "flightless" or "flying")

-country.end: whether  the species is endemic to a (value: "yes" or "no")