Received: 19 October 2022  | Revised: 5 January 2023  | Accepted: 11 January 2023 DOI: 10.1002/ece3.9775 R E S E A R C H A R T I C L E Implications for conservation assessment from flux in the botanical record over 20 years in southwest Ghana Cicely A. M. Marshall1  | Jonathan Dabo2 | Markfred Mensah2 | Patrick Ekpe3 | William D. Hawthorne4 1Department of Plant Sciences, University of Cambridge, Cambridge, UK Abstract 2Forestry Research Institute of Ghana, At best, conservation decisions can only be made using the data available at the time. Kumasi, Ghana For plants and especially in the tropics, natural history collections remain the best 3Ghana Herbarium, Department of Plant & Environmental Biology, University of available baseline information upon which to base conservation assessments, in spite Ghana, Legon, Ghana of well-d ocumented limitations in their taxonomic, geographic, and temporal cov- 4Department of Plant Sciences, University of Oxford, Oxford, UK erage. We explore the extent to which changes to the plant biological record over 20 years have changed our conception of the conservation importance of 931 plant Correspondence Cicely A. M. Marshall, Department of taxa, and 114 vegetation samples, recorded in forest reserves of the southwest Ghana Plant Sciences, University of Cambridge, biodiversity hotspot. 36% of species- level assessments changed as a result of new dis- Downing Street, Cambridge, CB23EA, UK. Email: cm997@cam.ac.uk tribution data. 12% of species accepted in 2016 had no assessment in 1996: of those, 20% are new species publications, 60% are new records for SW Ghana, and 20% Funding information Clarendon Fund; King's College are taxonomic resolutions. Apparent species ranges have increased over time as new Cambridge, University of Cambridge; records are made, but new species publications are overwhelmingly of globally rare Merton College, University of Oxford; Overseas Development Institute; Oxford species, keeping the balance of perceived rarity in the flora constant over 20 years. University Expeditions Council Thus, in spite of considerable flux at the species record level, range size rarity scores calculated for 114 vegetation samples of the reserves in 1996 and 2016 are highly correlated with each other: r(112) = 0.84, p < .0005, and showed no difference in mean score over 20 years: paired t(113) = −0.482, p = .631. This consistency in results at the area level allows for worthwhile conservation priority setting over time, and we argue is the better course of action than taking no action at all. K E Y W O R D S conservation, endemism, global plant inventory, species discovery, taxonomy, tropical biodiversity T A X O N O M Y C L A S S I F I C A T I O N Applied ecology, Biodiversity ecology, Botany, Conservation ecology, Ecosystem ecology, Global ecology, Landscape planning, Macroecology, Restoration ecology, Spatial ecology, Taxonomy This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. Ecology and Evolution published by John Wiley & Sons Ltd. Ecology and Evolution. 2023;13:e9775. www.ecolevol.org  | 1 of 12 https://doi.org/10.1002/ece3.9775 2 of 12  |     MARSHALL et al. 1  |  INTRODUC TION year, resulting in negative progress in real terms after the historical synonymy rate of 0.34% is applied. At best, conservation decisions can only be made using the data In Ghana, the Star system has been used since 1996 to prioritize available at the time. Short of surveying a whole region to make plant species for conservation (Hawthorne, 1996). In the Star sys- decisions locally, the best we can hope to do is to use all available tem, species are assigned to one of four categories of global range data and to characterize how our current data may be biasing our size rarity by reference to their area of occupancy (AOO) measured perspective. For plants, the most commonly used baseline infor- globally with a fixed cell size of 1 degree square (or 100 × 100 km mation upon which conservation assessments are based is the outside the tropics). Black Star species are the most globally rare geographic distribution of species in terms of area of occupancy species, with a mean global AOO of 2.7 degree squares. Gold Star (AOO) or extent of occupancy (EOO) as revealed by herbarium species have restricted ranges of 8 degree squares on average, Blue specimens (Brummitt, Bachman, Griffiths-L ee, et al., 2015). While Star species have a three times larger range of 24 degree squares, there have been significant efforts to digitize and curate herbar- while Green Star species are the most widespread. One of the te- ium records (Dauby et al., 2016; Enquist et al., 2016), the expense nets of Star rating is that all species should be rated using the best of collecting and storing specimens long- term has limited the information available at the time, even if this is perhaps incomplete, plant biological record in terms of its taxonomic, geographic, and with ratings being updated in light of new information if necessary temporal coverage and resolution (Meyer et al., 2016; Roberts (Hawthorne & Marshall, 2016). There is also no requirement to show et al., 2016; Schmidt- Lebuhn et al., 2013; Stropp et al., 2016). a decline in population size or distribution, as there is for the Red The principle problem with an incomplete baseline description List, which reduces the burden of proof further. All plant species in of plant species' distributions for conservation is that newly ac- tropical Africa have a published Star rating (Marshall et al., 2016). cessioned specimens (or new field observations) will tend to pro- Black Star species are legally protected from logging in Ghana: The duce increases in EOO and AOO (Brummitt, Bachman, Aletrari, Ghanaian Government's Timber Resources Management Act (Act et al., 2015). 547) and Timber Resources Management Regulations (LI 1649) of If our evidence threshold for conservation assessment is too 1998 underpin current forest regulations, and the enforced logging stringent, we fail to define a baseline against which to monitor manual is framed in the context of Star categories, where Black Star change for many species. If our evidence threshold is set too low, species are wholly protected from logging. we risk producing misleading or labile conservation assessments In Ghana, among other countries, area- based conservation ac- inappropriate for a legal framework. Protecting “undeserving” tion has also been taken on the basis of the Star ratings. Each Star species has economic costs, while failing to protect “deserving” category carries a weight in inverse proportion to its rarity. The species results in local or global extinctions. At the moment, our weights of the species present in a vegetation sample are averaged evidence threshold for conservation assessment via the IUCN Red to give a bioquality score for the vegetation overall, reflecting the List is set very high for plants. Authors need “good evidence to concentration of globally rare plants within the vegetation. In 1999, establish that a species is not undercollected” in order to pub- the Ghanaian Government established the High Forest Biodiversity lish an IUCN Red List assessment for a species with <5 records Conservation Project (HFBCP) funded by the World Bank, which (Brummitt, Bachman, Aletrari, et al., 2015), while 15 georefer- aimed to conserve biodiversity in these forests through the es- enced specimens per species are needed to produce Red List range tablishment and protection of Globally Significant Biodiversity estimates consistent with estimates based on all known specimens Areas (GSBAs). These reserves were identified and protected on (Rivers et al., 2011). Fifteen distinct georeferenced specimens are the basis of their high bioquality scores. This approach led to the a high bar for most plant species. It takes around 70 years from the establishment of 29 forest reserves amounting to c. 2300 km2 of collection of the first specimen to reach 15 specimens (Goodwin forest reserves or 13% of the total forest network and their ex- et al., 2020). 36.5% of the world's plant species (158,535 species) clusion from timber harvesting. In 2009, Ghana was the first are represented by five herbarium or field observations or fewer, country to sign a Forest Law Enforcement Governance and Trade with 28.3% (123,149 species) having three observations or fewer, (FLEGT) Voluntary Partnership Agreement (VPA) with the EU (The and 13.6% to 11.2% species with just one observation (Enquist European Community and the Republic of Ghana, 2010). As a con- et al., 2019). The result of this high burden of evidence is that Red dition of the VPA, it is stated that no Black Star species, as defined Listing progress for plants has been limited, at around 8.5% of ac- in Hawthorne, 1996, can be felled and that no timber can come cepted vascular plant species names, excluding those in need of from a GSBA. By 2007, the GSBAs showed improved afforestation updating (26,720 of 316,143). As much as 39% of plant species and rainfall patterns, reduced illegal tree felling and group hunting, selected randomly for inclusion in the Sampled Red List Index for less seasonal reduction of volumes of water bodies, and the cessa- Plants could not be evaluated against the Red List criteria, even tion of the use of poisonous chemicals in fishing (GEF Evaluation as Data Deficient (Brummitt, Bachman, Griffiths-L ee, et al., 2015). Office– UNDP Evaluation Office, 2007). For the 5 years preceding 2019, substantially fewer species were Hotspots of locally endemic or restricted range species are assessed for the Red List each year than were published the same treated as priority areas for plant conservation well beyond Ghana's 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MARSHALL et al.     |  3 of 12 GSBAs, for example, recently in the Key Biodiversity Areas designa- et al., 2016, 2022). The typical flora of the study area is lowland ev- tion (Brooks et al., 2006; IUCN, 2016a; Myers et al., 2000). There ergreen rainforest, with variations in species composition driven by has been some work addressing how uncertainty in our knowledge gradients in rainfall, disturbance, and local topology, with altitude of plant species' taxonomy and distribution manifests in lability of and historical climatic stability important at broader geographi- species' IUCN ratings over time (Goodwin et al., 2020; Lughadha cal scales (Marshall et al., 2022). Across the sampled sites, annual et al., 2019; Rivers et al., 2011). However, how incomplete knowl- precipitation averages 1910 mm/year, while mean temperatures av- edge at the species level manifests in conservation assessments of erage 26.5°C (Hijmans et al., 2005). Altitude above sea level has a areas is an important and underexplored question (Daru et al., 2020; minimum of 40 m, a maximum of 181 m, and a mean of 97 m. The Murray- Smith et al., 2009; Nelson et al., 1990). forest reserves of southwest Ghana are subject to anthropogenic Here, we describe changes in the nomenclature and documented disturbance, for example, accommodating logging concessions. distributions of the vascular plant species of the southwest Ghana forest reserves over 20 years (1996 to 2016). We explore the conse- quence of these changes for species-l evel conservation assessments 2.2  |  Field methods (Star ratings) and area-b ased conservation assessments (bioquality scores). We argue that in spite of considerable flux in the botani- Twenty-s even samples were enumerated in 2015 for this publica- cal record, area- based conservation assessments in particular have tion and were compiled with 87 botanic samples from the region remained remarkably stable over 20 years and have led to positive from different surveys (Table 1). The combined dataset contains outcomes for conservation in Ghana. By contrast, forests outside 12,232 records of 931 taxa identified to species level, from 114 of reserves experienced significant and growing threats from log- samples. With the exception of the Hall & Swaine, 1981 A plots, ging, fuelwood collection, agriculture, mining, and climate change which were 25 × 25 m samples of the vascular flora, and B plots (Aleman et al., 2018; Edwards et al., 2014; McClean et al., 2006). (the first 30– 60 vascular plant species encountered; Hall & Swaine, 1981), all samples were conducted using the Rapid Botanic Survey (RBS) method (Hawthorne & Marshall, 2016). RBS samples 2  |  MATERIAL S AND METHODS were bounded by survey effort rather than size, with the aim of re- cording all vascular plant species within a specified vegetation type 2.1  |  Study area and landscape unit, and enumerating at least 40 individual canopy trees. Identification and fieldwork was carried out by JD, MM, CM, Upper Guinea is a phytogeographical name for the forest zone of and WH, with assistance at the University of Ghana herbarium (GC) west Africa, running from Sierra Leone in the west to the Dahomey from PE. Permission to collect and export plants was obtained from Gap (Ghana) in the east, from the coast up to 350 km inland (Marshall the Ministry of Lands and Natural Resources, Samartex, and the et al., 2021; White, 1979). This area is included within the Western Wildlife Division for Ankasa RR. Specimens were collected of all African Forests ecoregion, which has been recognized as a bio- plants for which identification was not absolutely certain; these are diversity hotspot (Myers et al., 2000; Olson et al., 2001). The for- housed at the Daubeny herbarium, Oxford. The 114 sample loca- est reserves of SW Ghana, where this study is situated, have also tions (Figure 1) were situated in order to capture variation in forest been recognized as Key Biodiversity Areas (KBAs) (Key Biodiversity type, condition, and geography across the five forest reserves, as all Areas Partnership, 2022). Southwest Ghana is regarded as a hotspot the surveys had as their primary goal the purpose of baselining or of plant endemism within west Africa, being home to a high con- inventorying the forest reserves. Mapping was conducted in QGIS centration of globally rare species (Bongers et al., 2004; Marshall version 2.16.1. TA B L E 1 Datasets compiled for SW Ghana No. of samples: Boi Tano Jema Nini Dataset Reference/ownership Ankasa Tano Nimri Assemkron Suhien Total 2015RBS Marshall et al. (this publication) 2 16 9 0 0 27 2015TSP Marshall et al. (this publication) 3 0 0 0 0 3 HANDS Hall and Swaine (1981) 2 0 2 1 0 5 RBS1991 Hawthorne and Abu- Juam (1995), 7 4 4 3 0 18 Hawthorne (1996). Ghana Forestry Dept. ANK1998 Hawthorne (1998). Ghana Wildlife Dept. 45 0 0 0 10 55 GSBA Hawthorne (2002)/Ghana Biodiversity Unit 0 5 0 1 0 6 Ministry of Lands & Forestry 59 25 15 5 10 114 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 of 12  |     MARSHALL et al. F I G U R E 1 Map of 114 sample locations in the southwest Ghana study area, colored by GHI calculated using the Stars published in 2016 and including all species found. Reserves are named. 2.3  |  Conservation ratings over time by or shared with the last author (WH). In 2016, Star ratings for c. 40,000 African plant taxa were published from an updated curated 2.3.1  |  Species database describing the global range of tropical African plant spe- cies (Marshall et al., 2016). The method of assessment was the same, We use Star ratings as our species- level conservation assessment while the available data were different. (Hawthorne & Marshall, 2016; Marshall et al., 2016). Species are as- We examined the congruence between these 1996 and 2016 signed to one of four categories of range size rarity, measured by Star ratings, for 931 currently accepted plant species recorded at reference to their area of occupancy (AOO) measured globally with least once in 114 vegetation plots in southwest Ghana. All species a fixed cell size of 1 degree square (or 100 × 100 km outside the distribution records from 114 vegetation samples (Table 1) were tropics). Black Star species have a mean global AOO of 2.7 degree collected together in a single database table (see ‘Section 2.2’ for squares, Gold Star of 8, Blue of 24, and Green of 72. For species vegetation sampling detail). All records were analyzed at species or regions with poor coverage in terms of published degree square level, with subspecies and varieties reduced to species level, and any location records, proxies such as districts or prefecture- level oc- record identified to genus level or higher only also removed. All re- currence data have been used for assigning Stars. One degree (or cords were converted to their currently accepted name (current as 100 km) square is an appropriate grid size for analyzing plant species' of 2016), following the African Plants Database (Conservatoire et ranges at global or continental scope, reflecting the (lack of) spatial Jardin botaniques de la Ville de Genève and South African National resolution available in the herbarium record for most plant species Biodiversity Institute Pretoria, 2016). Two hundred and fifty spe- worldwide. cies names in the original records are now considered synonyms. Star ratings for 1403 species of the forest zone of Ghana were Nine of those records were then redundant, as they had been sunk published in 1996, using the best available distribution data avail- into names already listed (1 Black, 2 Gold, 2 Blue, and 4 Green). able at the time (Hawthorne, 1996). The basis of this was the range Homotypic synonyms with a merely cosmetic name change and no descriptions in the Flora of West Tropical Africa (FWTA second edi- associated range change are treated for this analysis as if they re- tion, 1952–1 972), along with field and herbarium records collected mained unchanged, that is, they are not included in the counts of 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MARSHALL et al.     |  5 of 12 TA B L E 2 Star ratings for all 931 currently accepted species changes or additions. Reddish Stars (S, R, P) and savanna (X) Stars recorded at least once in the 114 samples of the reserves of published in 1996 were excluded from the dataset and calculations, southwest Ghana. as these are no longer in use. A species' Star rating may change for two reasons: (i) new distri- BK GD BU GN Star (2016) (2016) (2016) (2016) bution data alters our understanding of the species' range; (ii) the species experiences a genuine increase or decrease in range size. BK (1996) 8 10 6 0 24 Star ratings may be added when (1) new species are published (and GD (1996) 0 18 80 7 105 found in southwest Ghana), including the splitting of taxa; (2) previ- BU (1996) 0 6 114 63 183 ously published species are newly recorded in (southwest) Ghana; GN (1996) 0 0 132 378 510 (3) “known unknowns” are resolved, that is, species which were con- Unknown 15 23 48 23 109 sidered to be of uncertain Star in 1996 are resolved to a Star in 2016 (1996) thanks to new sources of information. This flux is visualized with a 23 57 380 471 931 chord diagram implemented with R package circlize (Gu et al., 2014; Note: Star ratings in 1996 are compared with their destination rating R Core Team, 2020). in 2016. Column totals show the total number of species in each Star rating in 2016; row totals show the total number of species with each Star rating in 1996. To avoid double counting, species from 1996 which have since been synonymized to species already included are not 2.3.2  |  Areas repeated, so 24 rather than 25 BK star species of 1996 are referred to here (one Black Star species Trichoscypha chevalieri was sunk to T. We use the Genetic Heat Index (GHI) as our area-l evel conserva- lucens, already a Green Star species in 1996). tion rating (Hawthorne, 1996; Hawthorne & Marshall, 2016), the standard index of bioquality published in 1996 and used to des- 3  |  RESULTS ignate Ghana's GSBAs. GHI is a continuous metric representing the weighted global ranges (endemism) of the species present in 3.1  |  Stability of species-l evel conservation a plant community. The calculation is similar to range size rarity, assessments except that species' ranges are measured globally at one degree square resolution rather than within the study area at any resolu- Changes and additions to published Star ratings over 20 years are tion, species ranges are categorized into four groups, called Stars, explored for the forest zone of southwest Ghana (Table 2). All spe- rather than being treated continuously, and bioquality includes no cies recorded at least once in the reserves are considered, including measure of species richness, as is sometimes the case with range species which were not refound in 2015 and species which did not size rarity metrics. have a Star rating in 1996. Of 931 currently accepted species found GHI is calculated as the proportion of species belonging to in the 114 samples of the forest reserves of Ghana, half (51%) are each Star rating within a sample, where each species is inversely currently considered Green Star (the most widespread species) and weighted by the mean range size of its Star (BK = 27, GD = 9, 41% are Blue Star (the second most widespread rating). Just 2.5% BU = 3, GN = 0). The GHI for each sample was calculated from the of species are considered Black Star (the most range- restricted cat- species present, using the following formula, where NBK, NGD, egory), and 6% are Gold Star (Table 2). There has been considerable NBU, and NGN are the number of Black, Gold, Blue, and Green flux in the identity of the species present in each of the Star ratings Star species in a sample, and WBK, WGD, and WBU are the re- (Figure 2). spective weights. GHI = 100 × ((NBK ×WBK) + (NGD ×WGD) + (NBU ×WBU) + (NGN ×WGN))∕ (NBK + NGD + NBU + NGN) To investigate the change in these area- level conservation 3.1.1  |  Changes assessments over time, GHI was calculated for 114 vegetation samples using (i) the Stars published in 1996, and (ii) the Stars Considering only the species which had a known Star rating in 1996 published in 2016. Records identified to genus or family level only (species with no Star rating in 1996 are excluded), 36% of 1996 were dropped from the datasets. Analysis was conducted at the ratings are different now, while 64% of 1996 assessments have lowest named taxonomic level (species or named infraspecific remained stable (Table 3). Of the ratings which changed, a slightly taxa). Any record carrying a synonymous name was updated to the higher proportion of species were downgraded compared with up- accepted name before analysis, using the taxonomic framework graded: 20% of species ratings were downgraded, compared with of the African Plants Database. GHIs calculated as they appeared 16% of species ratings upgraded. in 1996, and 2016, are compared with each other via paired t- Commoner species have more stable Star ratings than rarer spe- test, Pearson correlation, and linear regression in base R (R Core cies: 74% of Green Star species have remained Green Star, compared Team, 2020). with 33% of Black Star species which have remained Black Star. For 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 6 of 12  |     MARSHALL et al. F I G U R E 2 Chord diagrams showing the conservation assessment trajectory of 931 currently accepted Ghanaian forest zone species recorded at least once in the reserves. (a) Chord and segment widths are proportional to the number of species in each category. (b) segment widths are scaled equally. TA B L E 3 Changes to the Star ratings published in 1996. The proportion of 1996 Star ratings which have remained stable (Star rating the same in 2016), have been upgraded (moved to a higher Star rating in 2016), or downgraded (to a lower Star category in 2016), grouped by original Star rating. Species with no Star rating in 1996 are excluded. BK (1996) GD (1996) BU (1996) GN (1996) Overall Stable 33.3% 17.1% 62.3% 74.1% 64% Upgraded – 0% 3.28% 25.9% 16% Downgraded 66.6% 82.9% 34.4% – 20% the rarest (Black Star) species, only one third of species considered New additions to the flora are disproportionately rare species: Black Star in 1996 remain Black Star species in 1996, with the re- A chi-s quare test of association shows that many more species maining two thirds downgraded to Gold or Blue Star species. A sim- additions proved to be Black or Gold Star species than would be ilar pattern of downgrading can be seen for Gold Star species and to expected given previously rated species, and many fewer than ex- a lesser extent Blue Star species. pected proved to be Green Star (Table 4). For the rarer Black and Downgrading of species from the rare categories to more wide- Gold Star species, additions contribute a significant proportion of spread categories is the result of increases in the documented range the currently accepted ratings: 65% of Black Star species were un- of species, due to more distribution data records being collected known in 1996, and 41% of Gold Star species were unknown in 1996. and becoming available online and for analysis. An inestimable pro- The majority of additions to the Star-r ated flora were the result portion of these records will represent genuine range extensions of new records for southwest Ghana or Ghana as a whole (60%); 20% for species, although the majority are very likely to be simply the are new species publications, and 20% are resolved names. Resolved result of better recording of established ranges. Overwhelmingly, names are the species whose distributions, or taxonomic status, the upgraded species are species which moved from Green to Blue were uncertain enough in 1996 to be given a Star rating of “?” at the Star: These are species which appeared to be widespread from their time. Many examples of new species publications (since 1996) are range description in FWTA (e.g., “to Lower Guinea”), but on consid- of species currently considered rare, for example, Pavetta abujuamii eration of their modern dot map distribution data have subsequently and P. ankasensis; Psychotria hawthornei, based on new collections; been shown to be sparse within that extent of occurrence. while others were based on reassessment of older specimens and went straight to Blue Star once named, for example, Eremospatha dransfieldii and Hypselodelphys triangularis. 3.1.2  |  Additions New records for the region include Of the 931 species now considered to be present in samples of southwest Ghana, 109 species (12%) had no Star rating in 1996 1. Species found in southwest Ghana for the first time since (Table 4). 1996, though previously recorded elsewhere in West Africa (e.g., 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MARSHALL et al.     |  7 of 12 TA B L E 4 New additions to the flora of southwest Ghana are overrepresented in the 2016 Black and Gold Star ratings and underrepresented in the 2016 Green Star ratings BK (2016) GD (2016) BU (2016) GN (2016) Species Star rating additions since 1996 15 (expected 2.7) 23 (expected 6.7) 48 (expected 45) 23 (expected 55) 109 Remainder (changes plus constant) 8 34 332 448 822 23 57 380 471 931 Note: Chi- square test of association, X- squared = 130.48, df = 3, p < .005. Tricalysia parva; Dichapetalum staudtii; Didelotia afzelii; Hunteria model was fitted to predict GHI scores in 2016 from GHI scores in simii; Mussaenda grandiflora; Pavetta subglabra; Pleioceras afzelii; 1996. A significant regression equation was found (F(1,112) = 260.45, and Whitfieldia colorata); p < .0005), with an adjusted R2 of 0.697. GHI in 2016 is equal to 2. Species not recorded as present in Ghana by the Flora of West 84.8 + 0.602*(GHI in 1996). Hotter samples have become cooler with Tropical Africa (FWTA) (e.g., Chassalia laxiflora (Liberia only); time, while cooler samples have become hotter with time. This is the Combretum fulvum (Guinea only); Crotonogyne caterviflora (Sierra result of changes to the overall number of Blue and Gold Star spe- Leone and Liberia only); and Crudia klainei (Ivory Coast, French cies in particular: cooler samples, dominated by Green Star species Cameroons, Gabon)); in 1996, have received an uplift in GHI as 26% of original Green Star 3. Species not mentioned at all by FWTA, being referred to there as species were upgraded to Blue Star species in 2016. Hotter samples, another currently recognized taxon or having been published be- with the highest concentrations of Black and Gold Star species, now tween FWTA publication and 1996, for example, Campylospermum have a reduced GHI as 83% of Gold Star species were downgraded laxiflorum/Ouratea laxiflora; Cissus miegei; Cleistanthus ripicola; (and were not replaced by new ratings, as Black Star species were). Cnestis bomiensis, Geophila flaviflora; and Laccosperma acutiflorum. In terms of conservation implications, we are concerned by both the magnitude and direction of change in the apparent conservation value of samples at the top end of the GHI scale, rather than the 3.1.3  |  Conservation priority species bottom. Using a score of 300 as a threshold for the hottest samples, of the 18 samples with scores above 300 in 1996, six samples remain Changes to the list of species considered to be of the highest conser- above 300 in 2016. The remaining 10 samples now fall below 300, vation priority are detailed in Table 5. Twenty- three Black star taxa are but still above 200, representing high conservation value. Only two currently accepted as present in SW Ghana. In 1996, there were 25 samples fell below 300 in 1996 but now have scores >300. Scores at Black Star species known from these reserves. Sixteen of those spe- the top end of the GHI scale have generally decreased slightly with cies considered Black Star in 1996 have been downgraded. One spe- time, and thus, fewer samples are not protected when they later cies has been sunk to a species already published in 1996 (Trichoscypha prove to warrant it. chevalieri to T. lucens, not included in calculations). Fifteen new Black Star species have been added to the list: 10 by new species publica- tions, 4 by new records in the region, and one species considered too 3.2  |  Inferred genuine changes in conservation uncertain to merit a Star rating in 1996 has been resolved as a Black status Star species. Eight Black star species are stable: recorded as Black Star species of the reserves in 1996, and still Black Star species of the re- The locations of the 2015 samples were not identical to any previ- serves in 2016. Stability of area- based conservation assessments. ous sample locations, and so we have no data to describe changes to The overall GHI for species of southwest Ghana calculated using the species composition of individual samples over time. Excepting the 1996 Stars is 218; the overall GHI for species of southwest Ankasa, these are production (logging) reserves with accessibility Ghana calculated using the 2016 Stars is very similar, at 216. On the dependent on which compartment is being logged at the time, and whole, the forces of change have converged to give the same overall the fate of individual samples over time would in any case depend impression of the balance of rarity and widespreadness in the flora. on whether or not it had just been logged or otherwise disturbed. This is reflected when we compare the GHIs for each sample cal- Neither was the most recent survey effort equal to all previous sur- culated using the Stars of 1996, to the GHIs for each sample calcu- vey efforts: Only 30 of the 114 samples considered were carried out lated using the Stars of 2016, via a paired Student's t-t est. GHIs were in 2015. However, we did work in the same reserves, choosing to normally distributed using both 1996 and 2016 Star ratings. The sample good condition samples of the apparent vegetation types in mean difference is negligible at −1.91 GHI points, and the 95% con- those same reserves (including in the GSBAs), using the same meth- fidence interval (−9.77 to 5.94) includes 0 (paired t(113) = −0.482, ods as in previous studies, so at the vegetation- type level, the data p = .631). Overall, the mean GHI score for the samples is the same, collected in 2015 comprise an informative resurvey effort. whether scores are calculated with the old Stars or the new Stars. Two hundred and eighty- six species were not refound in 2015, There is strong positive correlation between the 1996 and the out of a total of 931 species recorded at any time from these re- 2016 GHIs: r(112) = 0.84, p < .005 (Figure 3). A linear regression serves (30.7%). The 286 species which were not refound were not 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8 of 12  |     MARSHALL et al. TA B L E 5 Species currently or previously considered Black Star TA B L E 5 (Continued) species of southwest Ghana Star Star Star Star Species 2016 1996 Notes Species 2016 1996 Notes Shirakiopsis aubrevillei *VU BU BK Downgraded Calycobolus hallianus BK New species Synsepalum aubrevillei GD BK Downgraded Chytranthus verecundus BK BK Stable Trichoscypha chevalieri – BK Sunk to T. lucens Cola umbratilis BK BK Stable (GN) Dorstenia embergeri BK ? Resolved Note: Species denoted with asterisk * are species large enough trees to Ficus pachyneura BK New record be of potential interest for timber, but are nevertheless still protected 2015 from logging as they were designated Black Star in 1996. All are also Red Listed threatened species, EN, Endangered, VU, Vulnerable, and we Gaertnera luteocarpa BK New species do not recommend that their protection from logging is removed. (subsp. luteocarpa) Hymenostegia gracilipes BK BK Stable Leucomphalos discolor BK BK Stable Momordica sylvatica BK New species biased by 2016 Star (chi- square test of association, X- squared (df = 3, Monocyclanthus vignei BK BK Stable n = 931) = 1.91, p = .59). Nine Black Star species (Monocyclanthus Pavetta abujuamii BK New species vignei, Pavetta abujuamii, Pavetta sonjae, Pavetta subglabra, Psychotria Pavetta ankasensis BK New species longituba, Psychotria nigrostellata, Schefflerodendron ekpei, Suregada BK New species ivorensis, and Synsepalum ntimii) and 16 Gold Star species were not Pavetta sonjae refound: These are potentially of concern, but given the lower sam- Pavetta subglabra BK New record 1998 pling effort and the local rarity of these species we do not believe that Psychotria ankasensis BK BK Stable they are truly missing. Eighty-e ight species are unique to the 2015 dataset, that is, recorded for the first time in the reserves in 2015, Psychotria hawthornei BK New species including two Black Star species (Ficus pachyneura and Momordica Psychotria longituba BK BK Stable sylvatica) and four Gold Star species (9.45% of all species in the data- Psychotria nigrostellata BK New species set). These 88 species are also not biased by Star (chi-s quare test of Schefflerodendron ekpei BK New species association, X- squared (df = 3, n = 931) = 0.55, p- value = .91). Suregada ivorensis BK New record For all but the most widespread and well- collected plants of west 2015 Africa's forest zone, changes to our perception of species' taxonomy Synsepalum ntimii BK New species and distributions strongly outweigh our ability to detect real distri- Tapura ivorensis BK BK Stable bution changes over time. The likelihood of genuine and significant Tricalysia parva BK New record range extensions over 20 years for these species can be dismissed as 2015 low, given the ecology of the species and habitat changes ongoing in Alsodeiopsis chippii GD BK Downgraded the forest zone of west Africa (Stévart et al., 2019). Chrysophyllum GD BK Downgraded azaguieanum Dactyladenia hirsuta GD BK Downgraded 4  |  DISCUSSION Dasylepis blackii BU BK Downgraded Leptoderris cyclocarpa GD BK Downgraded This study showed that there has been considerable flux in our Leptoderris miegei GD BK Downgraded conceptions of taxonomy and distributions for plant species within Neolemonniera BU BK Downgraded southwest Ghana over 20 years, an area that is well studied relative clitandrifolia*EN to other tropical endemism hotspots. 36% of species changed global Nephthytis swainei GD BK Downgraded AOO category (Star category), and 12% of species occurrences in Pierreodendron kerstingii GD BK Downgraded our dataset have been contributed since 1996. Species tended to ap- *VU pear more widespread, rather than more restricted, over the 20-y ear Placodiscus bancoensis GD BK Downgraded re- evaluation period (20% vs 16%). This effect is most marked for the *VU globally rarest species, with two thirds of local endemics moving to Psychotria BU BK Downgraded a more widespread AOO category after 20 years and just one third brachyanthoides remaining in the same category. However, new species discoveries Psychotria subglabra GD BK Downgraded and new records for the region were disproportionately of globally Ruellia togoensis BU BK Downgraded rare species, so that the total number of local endemics known in Sclerosperma mannii BU BK Downgraded southwest Ghana was maintained over the 20 years. This explains 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MARSHALL et al.     |  9 of 12 F I G U R E 3 Correlation between GHIs calculated for 114 samples of southwest Ghana using the 2016 Stars, and the GHI of those same samples calculated using the 1996 Stars: r(112) = 0.84, p < .005. Gray diagonal line is the 1:1 line; black diagonal line is the best fit linear regression line; vertical and horizontal dashed lines show the samples with GHI >300 (“hotspots”) in 1996 and 2016, respectively. why mean endemism scores (bioquality scores) have remained con- principle is often touted in conservation science (IUCN, 2016b), but stant over 20 years, even though the species responsible for the it less often followed. When specimens represent the best available perception have changed. Importantly, new locally endemic species evidence for particular species, their use as a basis for extinction were found in the same broad locations as the now- downgraded risk assessment is appropriate, necessary, and urgent (Lughadha local endemics, in spite of random survey effort, so that the locali- et al., 2019). This case study supports the precautionary principle: ties which appeared as hotspots of endemism in 1996 still appeared The correct areas were given conservation funds despite undercol- that way in 2016. This stability in assessment for localities allows for lection limiting our knowledge of those plant species' distributions meaningful conservation priority setting over time and in the face of outside of the study area. Protected areas benefited from improved partial information. afforestation and rainfall patterns, reduced illegal tree felling and This lability in documented ranges and conservation assess- group hunting, less seasonal reduction of volumes of water bodies, ments is probably applicable to most groups of plants in the tropics. and the cessation of the use of poisonous chemicals in fishing (GEF In neotropical Myrcia, for example, around 33% of species changed Evaluation Office– UNDP Evaluation Office, 2007). their apparent EOO-b ased conservation status as the result of We proffer that it is not our 250 years of plant biological records 10 years of specimen collection, databasing, and taxonomic changes which are inappropriate for assessing plant conservation status, but (Nic Lughadha et al., 2019). Similar to our findings, transitions to rather the 22- year- old IUCN Red List system. Many authors have smaller EOO categories outnumbered transitions to larger EOO suggested approaches to accelerate Red Listing by offering more categories (8% vs 3%, respectively), and species previously deemed appropriate assessment alternatives for plants species. These focus data deficient transitioned to threatened status more often than to particularly around using coarser distribution data while relaxing not threatened (10% vs 7%, respectively; Nic Lughadha et al., 2019). the requirement for or using crude proxies for decline in habitat For Madagascan orchids, species described more recently have area, extent or quality (Darrah et al., 2017; Le Breton et al., 2019; smaller ranges and occupancies, fewer specimens and greater Marshall et al., 2016; Miller et al., 2012; Pelletier et al., 2018; Stévart perceived extinction risk status (Roberts et al., 2016). In the Cape et al., 2019). Another important nexus of work is to improve the Floristic Region, narrow-r ange taxa have constituted a significantly online mobilization and dissemination of published botanic records greater proportion of species discoveries since 1950 (Treurnicht from sources other than herbaria, including checklist data, Floras, et al., 2017). The almost universal lability of taxonomic nomencla- and scientific publications, into data standards and repositories ture has been perceived as awkward for conservation (Garnett & appropriate for conservation assessments (Agosti et al., 2022). As Christidis, 2017). Rather than indulge calls to “finalise” taxon names much as 39% of plant species selected randomly for inclusion in the (Thomson et al., 2018), accepting a degree of lability in conservation Sampled Red List Index for Plants could not be evaluated against assessments would be an alternative and more accurately reflect our the Red List criteria, even as Data Deficient (Brummitt, Bachman, evolving knowledge of the natural world. Griffiths- Lee, et al., 2015). A conservation assessment system that Although some relatively common species were previously con- cannot be applied to 39% of plant species, and likely the rarest sidered rare in our study, as a result of undersampling, no rare spe- and certainly the least well- documented 39% at that, is surely too cies were misclassified as common, so all the species that appeared stringent. The problem with applying too stringent a set of criteria worthy of attention at the time would have received it. Many rare relative to the available evidence is that we fail to define a base- species had been overlooked or were not yet described, and these line against which to monitor change for many species. Rather than were disproportionately found in original hotspot locations, despite accept that many plant species cannot be assessed and continue to equal resampling effort in cold and hot spots. The precautionary focus our efforts on only the most well-k nown and perhaps least 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 10 of 12  |     MARSHALL et al. threatened plant species, we support calls to assess plant species species and 100% of Guinea's Threatened Habitats would survive. using the evidence available and characterize how our current data Key Biodiversity Area assessment in Uganda showed that most of may be biasing our perspective, but ultimately to accept some labil- the remaining natural habitat was important for the conservation of ity in those assessments. globally and nationally threatened species and threatened habitat We recognize that the goal of the Red List is to assess extinc- (Plumptre et al., 2019). Emphasizing area- based assessments, such as tion risk in a comparable way across eukaryotes, rather than to as- range size rarity or weighted endemism, also reduces the importance sess a taxon's range (Collen et al., 2016). The relationship between of “fixing taxonomy” at the species level (Thomson et al., 2018). the risk of extinction and global range (endemism) is not absolute, Our study shows that locality assessments, which make use of the though narrow AOO or EOO is an all but essential prerequisite for summed (or averaged) signal from many species, were more stable a plant's assessment as threatened, and many restricted range spe- over time that the assessment of any one individual species. By act- cies are threatened (Robbirt et al., 2006). Tropical plant species with ing on the signal from the apparently rare species, some of which narrow ranges declared extinct are least likely to be discovered ex- later turned out to more widespread, conservation priority areas tant (Humphreys et al., 2019). We suggest that species known from were nevertheless identified and protected. It is surely not wise to <5 specimens should not be considered as unassessable, or Data imagine that we can ever finalize our estimates of AOO, EOO, Red Deficient. The precautionary principle, which is already enshrined List status, or nomenclature. Fortunately, the consequences of not in the IUCN guidance, should be employed in practice. Such species doing so may be less severe than we imagine. could be assessed by their AOO, along with number of locations, and appropriate proxies or evidence for a decline in habitat area, extent AUTHOR CONTRIBUTIONS or quality which could be relatively general in the case of plants. Our Cicely A. M. Marshall: Conceptualization (lead); data curation study suggests that such species may indeed prove to be more wide- (equal); formal analysis (lead); funding acquisition (equal); investi- spread (and less threatened) over time than they initially appeared. gation (equal); methodology (equal); project administration (equal); But in the meantime, their recognition and presence shines a light on resources (equal); writing – original draft (lead); writing – review an area which is poorly known and quite likely to harbor new records and editing (lead). Jonathan Dabo: Data curation (equal); investiga- or species new to science, which has the added benefit of being af- tion (equal). Markfred Mensah: Data curation (equal); investigation forded conservation protection, perhaps also helping to reduce the (equal). Patrick Ekpe: Data curation (equal); investigation (equal); historical 70 years required to collect 15 geolocated specimen local- resources (equal). William D. Hawthorne: Data curation (lead); fund- ities (Goodwin et al., 2020). ing acquisition (equal); investigation (equal); methodology (equal); Herbaria are principally a taxonomic repository, retaining type resources (equal); writing – review and editing (equal). specimens and significant range extensions such as new country re- cords. Vouchered specimens remain essential to provide verifiable ACKNOWLEDG MENTS identifications for work in ecology, phylogenetics, taxonomy, con- We thank the Oxford University Expeditions Council; Wilson Fund, servation, and pharmacology (Funk et al., 2018). A new sort of her- Department of Plant Sciences; and Merton College, University barium specializing in, or at least accepting, recollections of species of Oxford for funding the new fieldwork in SW Ghana. We thank from known localities, specimens located within the known EOO, Samartex, FORIG, and the Ghana Herbarium for facilitating the new sterile vouchers, and specimens which describe the species of a fieldwork. We thank Michael D. Swaine and J. D. Hall for permis- locality in depth would be useful for conservation assessment and sion to reuse their sample data. C.M. acknowledges support from reassessment. Such specimens need not be curated to such a high King's College, University of Cambridge and the Clarendon Fund, specification as traditional specimens or even be kept at all, once University of Oxford. W.H. acknowledges support from the ODA for photographed with a small amount of leaf material retained, for ex- previous fieldwork in SW Ghana. ample, for DNA extraction. Specimens could be linked via QR codes or similar to online platforms such as iNaturalist, providing supple- CONFLIC T OF INTERE S T mentary access to living plant photographs, identification history, lo- The authors declare no competing interests. cation maps, and potentially additional geolocated records from the area, without increasing physical storage requirements (Heberling & DATA AVAIL ABILIT Y S TATEMENT Isaac, 2018). The data that support the findings of this study, including all plot We further suggest a greater emphasis is placed on area-b ased data and species data, are publically available at Marshall, Cicely conservation assessment than species-l evel assessment (Plumptre et al. (2023), Data for Implications for conservation assessment et al., 2019). Habitat assessment is an extension from species-l evel from flux in the botanical record over 20 years in southwest Ghana, assessment, an approach adopted in the Star and bioquality system, Dryad, Dataset, https://doi.org/10.5061/dryad.qbzkh1 8n8. as well as Key Biodiversity Areas and Tropical Important Plant Areas (TIPAs), for example, in Guinea (Couch et al., 2019). If all 22 Guinean ORCID TIPAs could be protected, over 60% of Guinea's threatened plant Cicely A. M. Marshall https://orcid.org/0000-0002-7397-6472 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MARSHALL et al.     |  11 of 12 R E FE R E N C E S Enquist, B. J., Feng, X., Boyle, B., Maitner, B., Newman, E. A., Jørgensen, Agosti, D., Benichou, L., Penev, L., & Hyam, R. (2022). A possible work- P. M., Roehrdanz, P. R., Thiers, B. M., Burger, J. R., Corlett, R. T., flow from new and legacy publications to keep the world Flora on- Couvreur, T. L. P., Dauby, G., Donoghue, J. C., Foden, W., Lovett, J. line up to date with new species and augmenting taxonomic treat- C., Marquet, P. A., Merow, C., Midgley, G., Morueta- Holme, N., … ments. Biodiversity Information Science and Standards, 6, e91241. McGill, B. J. (2019). The commonness of rarity: Global and future https://doi.org/10.3897/biss.6.91241 distribution of rarity across land plants. Science Advances, 5, 1– 14. Aleman, J. C., Jarzyna, M. A., & Staver, A. C. (2018). Forest extent and https://doi.org/10.1126/sciadv.aaz0414 deforestation in tropical Africa since 1900. Nature Ecology and Funk, V. A., Edwards, R., Keeley, S., Funk, V. A., Edwards, R., & Keeley, S. Evolution, 2, 26– 33. https://doi.org/10.1038/s41559 -0 17-0 406-1 (2018). The problem with(out) vouchers. Taxon, 67, 3– 5. Bongers, F., Poorter, L., & Hawthorne, W. D. (2004). The forests of upper Garnett, S. T., & Christidis, L. (2017). Taxonomy anarchy hampers conser- Guinea: Gradients in large species composition. In L. Poorter, F. vation. Nature, 546, 25–2 7. https://doi.org/10.1038/546025a Bongers, F. N Kouamé & W. D. Hawthorne (Eds.), Biodiversity of GEF Evaluation Office–U NDP Evaluation Office, (2007). https://sgp. West African Forests. CABI Publishing, UK. undp.org/case- studi es- 189/253-g hana - sgp-c ase- study.html Brooks, T. M., Mittermeier, R. A., Da Fonseca, G. A. B., Gerlach, J., Goodwin, Z. A., Muñoz- Rodríguez, P., Harris, D. J., Wells, T., Wood, J. Hoffmann, M., Lamoreux, J. F., Mittermeier, C. G., Pilgrim, J. D., & R. I., Filer, D., & Scotland, R. W. (2020). How long does it take to Rodrigues, A. S. L. (2006). Global biodiversity conservation priori- discover a species? Systematics and Biodiversity, 0, 1–1 0. https://doi. ties. Science, 313, 58– 61. https://doi.org/10.1126/scien ce.1127609 org/10.1080/147720 00.2020.1751339 Brummitt, N. A., Bachman, S. P., Aletrari, E., Chadburn, H., Griffiths-L ee, Gu, Z., Gu, L., Eils, R., Schlesner, M., & Brors, B. (2014). Circlize imple- J., Lutz, M., Moat, J., Rivers, M. C., Syfert, M. M., & Nic Lughadha, E. ments and enhances circular visualization in R. Bioinformatics, 30, M. (2015). The sampled red list index for plants, phase ii: Ground- 2811–2 812. https://doi.org/10.1093/bioinf ormat ics/btu393 truthing specimen- based conservation assessments. Philosophical Hall, J. B., & Swaine, M. D. (1981). Distribution and ecology of vascular plants Transactions of the Royal Society B Biological Science, 370, 1– 11. in a tropical rainforest. Forest vegetation in Ghana. Junk Publishers. https://doi.org/10.1098/rstb.2014.0015 Hawthorne, W. D. (1996). Holes and the sums of parts in Ghanaian forest: Brummitt, N. A., Bachman, S. P., Griffiths-L ee, J., Lutz, M., Moat, J. Regeneration, scale and sustainable use. Proceedings of the Royal F., Farjon, A., Donaldson, J. S., Hilton-T aylor, C., Meagher, T. R., Society of Edinburgh, Section B: Biological Sciences, 104, 75– 176. Albuquerque, S., Aletrari, E., Andrews, A. K., Atchison, G., Baloch, Hawthorne, W. D. (1998). Plants in Ankasa, Nini- Suhien, and Bia. Review E., Barlozzini, B., Brunazzi, A., Carretero, J., Celesti, M., Chadburn, of existing knowledge, results from a new survey and recommen- H., … Lughadha, E. M. N. (2015). Green plants in the red: A baseline dations for management plans. Protected Areas Development global assessment for the IUCN sampled red list index for plants. Programme. Western Region, Ghana. ULG Consultants LTD in PLoS One, 10, 1–2 2. https://doi.org/10.1371/journ al.pone.0135152 Association with S. A. Agrer, N.V. Collen, B., Dulvy, N. K., Gaston, K. J., Gärdenfors, U., Keith, D. A., Punt, A. Hawthorne, W. D. (2002). Final report of the floral survey of the biodiver- E., Regan, H. M., Böhm, M., Hedges, S., Seddon, M., Butchart, S. H. sity component of the NRMP. Forestry Commission, Biodiversity M., Hilton- Taylor, C., Hoffmann, M., Bachman, S. P., & Akçakaya, H. Conservation Component. Ministry of Lands and Forestry, Ghana. R. (2016). Clarifying misconceptions of extinction risk assessment Hawthorne, W. D., & Abu-J uam, M. (1995). Forest Protection in Ghana. with the IUCN red list. Biology Letters, 12, 20150843. https://doi. IUCN/ODA/Forest Department Republic of Ghana xvii + 203 pp. org/10.1098/rsbl.2015.0843 Hawthorne, W. D., & Marshall, C. A. M. (2016). A manual for rapid botanic Conservatoire et Jardin botaniques de la Ville de Genève and South survey (RBS) and measurement of vegetation bioquality. African National Biodiversity Institute Pretoria. (2016). African Heberling, J. M., & Isaac, B. L. (2018). iNaturalist as a tool to expand the Plant Database (version 3.4.0). http://www.ville - ge.ch/musin fo/ research value of museum specimens. Applications in Plant Sciences, bd/cjb/afric a/ 6, 1–8 . https://doi.org/10.1002/aps3.1193 Couch, C., Cheek, M., Haba, P., Molmou, D., Williams, J., Magassouba, Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G., & Jarvis, A. (2005). S., Doumbouya, S., & Yaya Diallo, M. (2019). Threatened habitats Very high resolution interpolated climate surfaces for global land and tropical important plant areas (TIPAs) of Guinea, West Africa. areas. International Journal of Climatology, 25, 1965– 1978. https:// Darrah, S. E., Bland, L. M., Bachman, S. P., Clubbe, C. P., & Trias-B lasi, doi.org/10.1002/joc.1276 A. (2017). Using coarse- scale species distribution data to predict Humphreys, A. M., Govaerts, R., Ficinski, S. Z., Nic Lughadha, E., & extinction risk in plants. Diversity and Distributions, 23, 435–4 47. Vorontsova, M. S. (2019). Global dataset shows geography and https://doi.org/10.1111/ddi.12532 life form predict modern plant extinction and rediscovery. Nature Daru, B. H., Farooq, H., Antonelli, A., & Faurby, S. (2020). Endemism Ecology and Evolution, 3, 1043–1 047. https://doi.org/10.1038/ patterns are scale dependent. Nature Communications, 11, 1–1 1. s41559 - 019- 0906- 2 https://doi.org/10.1038/s4146 7- 020- 15921- 6 IUCN. ( 2016a). A global standard for the identification of key biodiver- Dauby, G., Zaiss, R., Blach- Overgaard, A., Catarino, L., Damen, T., sity areas, Version 1.0. Deblauwe, V., Dessein, S., Dransfield, J., Droissart, V., Duarte, M. IUCN Standards and Petitions Subcommittee. (2016b). Guidelines C., Engledow, H., Fadeur, G., Figueira, R., Gereau, R. E., Hardy, O. J., for Using the IUCN Red List Categories and Criteria. Version 12. Harris, D. J., De Heij, J., Janssens, S., Klomberg, Y., … Couvreur, T. Prepared by the Standards and Petitions Subcommittee. http:// L. P. (2016). RAINBIO: A mega- database of tropical African vascular www.iucnre dlist.org/docume nts/RedLi stGuid elin es.pdf plants distributions. PhytoKeys, 74, 1– 18. https://doi.org/10.3897/ Key Biodiversity Areas Partnership. (2022). Key Biodiversity Areas fact- phytok eys.74.9723 sheet: Ankasa Resource Reserve – Nini- Sushien National Park. Edwards, D. P., Sloan, S., Weng, L., Dirks, P., Sayer, J., & Laurance, W. F. Extracted from the World Database of Key Biodiversity Areas. (2014). Mining and the African environment. Conservation Letters, 7, Developed by the Key Biodiversity Areas Partnership: BirdLife 302– 311. https://doi.org/10.1111/conl.12076 International, IUCN, American Bird Conservancy, Amphibian Enquist, B., Condit, R., Peet, R., Schildhauer, M., & Thiers, B. (2016). Survival Alliance, Conservation International, Critical Ecosystem Cyberinfrastructure for an integrated botanical information Partnership Fund, Global Environment Facility, Global Wildlife network to investigate the ecological impacts of global climate Conservation, NatureServe, Rainforest Trust, Royal Society for the change on plant biodiversity. https://doi.org/10.7287/peerj. Protection of Birds, World Wildlife Fund and Wildlife Conservation prepr ints.2615 Society. http://www.keybi odive rsitya reas.org/ 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 12 of 12  |     MARSHALL et al. Le Breton, T. D., Zimmer, H. C., Gallagher, R. V., Cox, M., Allen, S., & Auld, Plumptre, A. J., Ayebare, S., Behangana, M., Forrest, T. G., Hatanga, P., T. D. (2019). Using IUCN criteria to perform rapid assessments of Kabuye, C., Kirunda, B., Kityo, R., Mugabe, H., Namaganda, M., at-r isk taxa. Biodiversity and Conservation, 28, 863– 883. https://doi. Nampindo, S., Nangendo, G., Nkuutu, D. N., Pomeroy, D., Tushabe, org/10.1007/s10531 - 019-0 1697- 9 H., & Prinsloo, S. (2019). Conservation of vertebrates and plants in Lughadha, E. N., Walker, B. E., Canteiro, C., Chadburn, H., Davis, A. P., Uganda: Identifying key biodiversity areas and other sites of na- Hargreaves, S., Lucas, E. J., Schuiteman, A., Williams, E., Bachman, tional importance. Conservation Science Practice, 1, 1–1 2. https:// S. P., Baines, D., Barker, A., Budden, A. P., Carretero, J., Clarkson, J. doi.org/10.1111/csp2.7 J., Roberts, A., & Rivers, M. C. (2019). The use and misuse of her- R Core Team. (2020). R: A language and environment for statistical comput- barium specimens in evaluating plant extinction risks. Philosophical ing. R Foundation for Statistical Computing. Transactions of the Royal Society B, 374, 20170402. https://doi. Rivers, M. C., Taylor, L., Brummitt, N. A., Meagher, T. R., Roberts, D. L., org/10.1098/rstb.2017.0402 & Lughadha, E. N. (2011). How many herbarium specimens are Marshall, C. A. M., Dabo, J., Mensah, M., Ekpe, P., Kpadehyea, J. T., needed to detect threatened species? Biological Conservation, 144, Haba, O. O., Bilivogui, D., & Hawthorne, W. D. (2022). Predictors 2541–2 547. https://doi.org/10.1016/j.biocon.2011.07.014 of plant endemism in two west African forest hotspots. Frontiers Robbirt, K. M., Roberts, D. L., & Hawkins, J. A. (2006). Comparing IUCN Ecology and Evolution, 10, 980660. https://doi.org/10.3389/ and probabilistic assessments of threat: Do IUCN red list criteria fevo.2022.980660 conflate rarity and threat? Biodiversity and Conservation, 15, 1903– Marshall, C. A. M., Wieringa, J. J., & Hawthorne, W. D. (2016). Bioquality 1912. https://doi.org/10.1007/s10531 - 005-4 307- 2 hotspots in the tropical African Flora. Current Biology, 26, 3214– Roberts, D. L., Taylor, L., & Joppa, L. N. (2016). Threatened or data defi- 3219. https://doi.org/10.1016/j.cub.2016.09.045 cient: Assessing the conservation status of poorly known species. Marshall, C. A. M., Wieringa, J. J., & Hawthorne, W. D. (2021). An in- Diversity and Distributions, 22, 558– 565. https://doi.org/10.1111/ terpolated biogeographical framework for tropical Africa using ddi.12418 plant species distributions and the physical environment. Journal of Schmidt- Lebuhn, A. N., Knerr, N. J., & Kessler, M. (2013). Non- geographic Biogeography, 48, 23– 36. https://doi.org/10.1111/jbi.13976 collecting biases in herbarium specimens of Australian daisies McClean, C. J., Doswald, N., Küper, W., Sommer, J. H., Barnard, (Asteraceae). Biodiversity and Conservation, 22, 905–9 19. https:// P., & Lovett, J. C. (2006). Potential impacts of climate doi.org/10.1007/s10531 -0 13-0 457-9 change on sub- Saharan African plant priority area selec- Stévart, T., Dauby, G., Lowry, P., Blach- Overgaard, A., Droissart, V., tion. Diversity and Distributions, 12, 645– 655. https://doi. Harris, D. J., Mackinder, A. B., Schatz, G. E., Sonké, B., Sosef, M. org/10.1111/j.1472-4 642.2006.00290.x S. M., Svenning, J. C., Wieringa, J., & Couvreur, T. L. P. (2019). A Meyer, C., Weigelt, P., & Kreft, H. (2016). Multidimensional biases, gaps third of the tropical African flora is potentially threatened with ex- and uncertainties in global plant occurrence information. Ecology tinction. Science Advances, 5, eaax9444. https://doi.org/10.1126/ Letters, 19, 992– 1006. https://doi.org/10.1111/ele.12624 sciadv.aax9444 Miller, J. S., Porter-M organ, H. A., Stevens, H., Boom, B., Krupnick, Stropp, J., Ladle, R. J., Ana, A. C., Hortal, J., Gaffuri, J., Temperley, G. A., Acevedo- Rodríguez, P., Fleming, J., & Gensler, M. (2012). H., Olav Skøien, J., & Mayaux, P. (2016). Mapping ignorance: Addressing target two of the global strategy for plant conservation 300 years of collecting flowering plants in Africa. Global Ecology by rapidly identifying plants at risk. Biodiversity and Conservation, and Biogeography, 25, 1085–1 096. https://doi.org/10.1111/ 21, 1877–1 887. https://doi.org/10.1007/s1053 1- 012- 0285- 3 geb.12468 Murray- Smith, C., Brummitt, N. A., Oliveira- Filho, A. T., Bachman, S., Moat, The European Community and the Republic of Ghana (2010). https:// J., Lughadha, E. M. N., & Lucas, E. J. (2009). Plant diversity hotspots eur- lex.europa.eu/legal -c onte nt/EN/ALL/?uri=CELEX %3A220 in the Atlantic coastal forests of Brazil. Conservation Biology, 23, 10A031 9%2801%29 151–1 63. https://doi.org/10.1111/j.1523- 1739.2008.01075.x Thomson, S. A., Pyle, R. L., Ahyong, S. T., Alonso- Zarazaga, M., Ammirati, Myers, N., Mittermeier, R. A., Mittermeier, C. G., Kent, C. A. B., & Kent, J., Araya, J. F., Ascher, J. S., Audisio, T. L., Azevedo- Santos, V. M., J. (2000). Biodiversity hotspots for conservation priorities. Nature, Bailly, N., Baker, W. J., Balke, M., Barclay, M. V. L., Barrett, R. L., 403, 853– 858. Benine, R. C., Bickerstaff, J. R. M., Bouchard, P., Bour, R., Bourgoin, Nelson, B. W., Ferreira, C. A. C., Da Silva, M. F., & Kawasaki, M. L. T., … Zhou, H.- Z. (2018). Taxonomy based on science is necessary (1990). Endemism centres, refugia and botanical collection den- for global conservation. International Commission on Zoological sity in Brazilian Amazonia. Nature, 345, 714– 716. https://doi. Nomenclature, 45, 117. org/10.1038/345714a0 Treurnicht, M., Colville, J. F., Joppa, L. N., Huyser, O., & Manning, J. Nic Lughadha, E. M., Graziele Staggemeier, V., Vasconcelos, T. N. C., (2017). Counting complete? Finalising the plant inventory of a global Walker, B. E., Canteiro, C., & Lucas, E. J. (2019). Harnessing the po- biodiversity hotspot. PeerJ, 2017, 1–1 1. https://doi.org/10.7717/ tential of integrated systematics for conservation of taxonomically peerj.2984 complex, megadiverse plant groups. Conservation Biology, 33, 511– White, F. (1979). The Guineo-C ongolian region and its relationships to 522. https://doi.org/10.1111/cobi.13289 other Phytochoria. Bulletin du Jardin Botanique National de Belgique, Olson, D. M., Dinerstein, E., Wikramanayake, E. D., Burgess, N. D., 49, 11. https://doi.org/10.2307/3667815 Powell, G. V. N., Underwood, E. C., D'amico, J. A., Itoua, I., Strand, H. E., Morrison, J. C., Loucks, C. J., Allnutt, T. F., Ricketts, T. H., Kura, Y., Lamoreux, J. F., Wettengel, W. W., Hedao, P., & Kassem, K. R. (2001). Terrestrial ecoregions of the world: A new map of How to cite this article: Marshall, C. A. M., Dabo, J., Mensah, life on earth. Bioscience, 51, 933. https://doi.org/10.1641/0006- 3568(2001)051[0933:teotw a]2.0.co;2 M., Ekpe, P., & Hawthorne, W. D. (2023). Implications for Pelletier, T. A., Carstens, B. C., Tank, D. C., Sullivan, J., & Espíndola, A. conservation assessment from flux in the botanical record (2018). Predicting plant conservation priorities on a global scale. over 20 years in southwest Ghana. Ecology and Evolution, 13, Proceedings of the National Academy of Sciences of the United e9775. https://doi.org/10.1002/ece3.9775 States of America, 115, 13027– 13032. https://doi.org/10.1073/ pnas.18040 98115 20457758, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ece3.9775 by University of Ghana - Accra, Wiley Online Library on [18/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License