Policy Research Working Paper 11045 Implementing 30x30 Lessons from Country Case Studies Susmita Dasgupta Brian Blankespoor David Wheeler Development Research Group A verified reproducibility package for this paper is Development Data Group & available at http://reproducibility.worldbank.org, Environment Global Department click here for direct access. January 2025 Policy Research Working Paper 11045 Abstract The publication of nearly 600,000 new species occurrence Unlike previous efforts, this approach assigns equal weight maps using Global Biodiversity Information Facility data to all vertebrates, invertebrates, plants, and other species provides an opportunity to reassess international species mapped in the database. A spatially efficient algorithm iden- protection with broader representation for plants, inver- tifies priority localities for establishing new protected areas tebrates, and other species. This development aligns with to safeguard unprotected species. The findings reveal that the global 30x30 initiative, where 188 governments have initial conditions, such as existing protection levels and the committed to expanding terrestrial and marine protection spatial clustering of unprotected species, greatly influence to cover 30 percent of the planet by 2030. This study lever- outcomes. Unprotected species are shown to be spatially ages Global Biodiversity Information Facility occurrence clustered in some countries but not in others, and the rep- maps to identify new opportunities for species protection resentation of different taxa among unprotected species is in 10 countries in Latin America (Brazil, Costa Rica, and found to vary significantly across countries. Some countries Ecuador), Africa (Cameroon, South Africa, and Madagas- can achieve full protection within the 30 percent territorial car), and the Asia-Pacific region (Papua New Guinea, the limit, while others may need to exceed it. However, in all Philippines, India, and China). By focusing on individual cases, spatial clustering enables significant protection gains countries, the paper emphasizes the importance of local through modest expansions of protected areas, demonstrat- conservation stewardship. Both terrestrial and marine cases ing a path forward for enhancing biodiversity conservation are analyzed, with particular attention to endemic species. within global commitments. This paper is a product of the Development Research Group and the Development Data Group, Development Economics and the Environment Global Department. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at sdasgupta@worldbank. org. A verified reproducibility package for this paper is available at http://reproducibility.worldbank.org, click here for direct access. RESEA CY LI R CH PO TRANSPARENT ANALYSIS S W R R E O KI P NG PA The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Implementing 30x30: Lessons from Country Case Studies Susmita Dasgupta Brian Blankespoor David Wheeler Funding: This research was funded by the Global Environment Facility. Acknowledgements: We acknowledge the use of georeferenced species occurrence data provided by GBIF, which was downloaded following their citation guidelines (https://www.gbif.org/citation-guidelines#derivedDatasets). The GBIF occurrence data was accessed via Google BigQuery on February 17, 2024. We extend our gratitude to Pritthijit (Raja) Kundu for his valuable assistance with the graphics. findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Keywords: Biodiversity Conservation, Protected Areas, Endemic Species, Kunming-Montreal Global Biodiversity Framework JEL classification: Q57 1. Introduction The world is rapidly losing biodiversity. Pimm et al. (2014)1 found that the current rate of species extinction is at least 1,000 times the background rate. Supporting evidence from the Living Planet Index (LPI), which tracks population trends for vertebrate species in terrestrial, freshwater, and marine habitats, shows a 69% decline since 1970. The LPI informs the Convention on Biological Diversity (CBD) and its Conference of Parties (COP). In response to indicators of rapid decline, 188 governments ratified the Kunming-Montreal Global Biodiversity Framework (GBF) at COP 15 in December 2022. Among other measures, the Framework commits participants to protecting 30% of the planet by 2030. This has been labeled the Global 30x30 Initiative and commonly abbreviated as “30x30”. Effective implementation of 30x30 requires addressing two key questions: (1) What is the spatial distribution of global biodiversity that should be protected? (2) How can protecting 30% of the planet best conserve this biodiversity? In a previous paper (Dasgupta et al. 2024), we addressed the first question using the Global Biodiversity Information Facility (GBIF), which has expanded to include occurrences for over 2 million species. In the past two years, the GBIF has added about 1.3 million occurrence records daily. Most records include locational coordinates, enabling new estimates of spatial distributions for previously unmapped species and improved estimates for species with existing maps. Using machine-based pattern recognition, we estimated spatial occurrence maps for over 600,000 species. These maps complement previous work by greatly expanding representation for plants, invertebrates and other non-vertebrate species. Our algorithm allows rapid updates and new maps as the GBIF data increase. In this paper, we draw on the previous work to explore some implications for 30x30 in a sample of 10 countries in Latin America, Africa and the Asia-Pacific region. We focus on individual countries to highlight the role of conservation stewardship in local settings. Our country analyses begin with current protected areas, which exhibit great cross-country variation in territorial coverage and representation of species taxa. We consider both terrestrial and marine cases, focusing on species that are endemic to each country. Drawing on our GBIF spatial database for nearly 600,000 species, we overlay species occurrence maps on the most recent protected-area maps from the World Database of Protected Areas. This enables identification of endemic species whose coverage by existing protected areas is either non-existent or extremely sparse. Our approach departs from many previous exercises by giving equal weight to all vertebrates, invertebrates, plants, and other species whose occurrence regions are mapped in our database. We use a spatially-efficient algorithm to identify a hierarchy of priorities for localities where new protected areas would offer coverage to unprotected species. Our country cases draw on our results to explore the spatial implications for countries’ 30x30 commitments. 1 Pimm, S. L., Jenkins, C. N., Abell, R., Brooks, T. M., Gittleman, J. L., Joppa, L. N., ... & Sexton, J. O. 2014. The biodiversity of species and their rates of extinction, distribution, and protection. Science, 344(6187), 1246752. 2 Rather than presenting a detailed description of our methodology at the outset, we use our country cases to introduce its features sequentially. We begin with an econometric analysis that identifies the appropriate scale for envisioned protected areas in each country. 2. Country Size and Protected Area Scale In each country, our sequential algorithm begins by overlaying the occurrence maps of all unprotected species on a spatial grid. It counts resident species for each grid cell and identifies the highest-count cell (P1) as the priority 1 candidate for protection coverage. In step 2, it sets aside the species resident in P1 and repeats the grid cell count for the remaining species. It identifies the highest-count cell as P2. The algorithm can repeat this exercise until all species have coverage in at least one grid cell.2 A critical feature of our approach is the selection of grid scale for each country. We use a spatial grid whose cell area is the same as the average protected-area size, controlling for country size. Figure 1 displays the relationship (in logarithms) between country area and mean protected area (PA) size in sq km for 212 countries and political units.3 The regression results in Table 1 confirm the strong visual relationship: Mean PA size increases by 0.614% with each 1% increase in country area, with very high significance (regression t-statistic of 15.474). Using this result, we can predict the mean PA size for any area and translate it to an appropriate grid cell resolution.5 2 When maximum resident populations are the same in more than one cell, the algorithm makes a random selection. 3 We compute average protected area size from WDPA (2024) and country area from the World Bank’s country shapefile (World Bank, 2024). 4 The t-statistic is the estimated regression coefficient divided by its standard error. Classical significance with 95% confidence generally requires a t-statistic above 2.0. 5 We could compute mean PA size directly for individual countries, but across the 212 countries and political units we observe great variation in both the numbers and sizes of PAs. We believe that our logarithmic regression provides more robust and stable results for the computation of appropriate grid scales. 3 Figure 1: Mean protected-area size vs. country area Table 1: Regression results (1) VARIABLES Log Mean Protected Area Size (sq.km.) Log Country Area (sq.km.) 0.614*** (0.0397) Constant -2.587*** (0.442) Observations 212 R-squared 0.533 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 4 3. Priority Area Selection: Brazil and Cameroon We introduce our methodology with applications for Brazil and Cameroon, whose size difference dictates grid cell resolutions of 0.332 and 0.136 decimal degrees (dd) respectively. With our algorithm, each step identifies a new PA in a grid cell with side dimensions of 0.332 dd (36.9 km) in Brazil, and 0.136 dd (15.1 km) in Cameroon. Brazil Figure 3 displays Brazil’s terrestrial protected areas in dark green. Its existing protected areas cover about 2.6 million sq km, or 30.6% of the national land territory (8.5 million sq km). Brazil has 20,245 endemic species in our database, and its current protection system is impressive by international standards. Brazil’s 5,499 protected areas incorporate 30.6% of its land area and provide significant coverage for 93% of its endemic species in our database. At the same time, 1,412 endemic species have occurrence areas that either lie totally outside existing protected areas or have such sparse coverage that they are essentially unprotected.6 To illustrate, Figure 2 displays images of some unprotected endemic species and Figure 3 displays occurrence maps and images of some unprotected species.7 The figure includes vertebrates (Figure 3a), plants (3b) and arthropods (3c). The public-domain requirement for our images has produced a quasi-random selection of species for Figure 2, although their small number cannot ensure that they are representative of Brazil’s 1,412 unprotected species. However, two striking features are evident: Many species are in eastern and southeastern Brazil, and the majority have relatively small ranges. These features are logical from a spatial perspective. First, protected areas are generally much larger in the north and northwest, so more local species are protected there. Second, protected areas are sufficiently pervasive in Brazil that species with large occurrence areas will often have significant protection in parts of their ranges. It is also worth noting that many unprotected species in our database have been mapped for the first time by Dasgupta et al. (2024), and many are in groups (e.g. arthropods) that have not received primary attention in protected area selection. Overall, Figure 3 suggests that these species have a significant presence in eastern and southeastern regions where protected areas are substantially smaller than in the north and northwest. 6 We define sparse coverage as protected-area coverage which is less than 5% of a species’ occurrence area and less than 25 sq km. 7 This paper contains many high-resolution species images because we believe that understanding of the protection problem can be enhanced by making unprotected species more visible. All images are in the public domain, and we have devoted considerable effort to verifying that they correctly represent individual species. However, we have no field expertise in this area, and we recognize that some expert readers may well recognize discrepancies. We welcome suggestions for improvement of our image presentation, with the proviso that all images must be in the public domain. 5 Figure 2: Brazil – Images of some unprotected species 6 Figure 3: Brazil - Unprotected species 3a: Vertebrates 3b: Plants 7 3c: Arthropods The four panels of Figure 4 illustrate the iterative steps that identify the four highest-priority areas for new protection status in Brazil. As previously noted, the grid scale for Brazil is 0.332 dd (about 36.9 km on a side). For the 1,412 unprotected species, Figure 4a reports initial counts by grid cell. Cell counts are color-coded from light yellow to red and dark brown, which identify the highest counts. Figure 4a shows many significant clusters in eastern and southeastern Brazil, with the maximum cell count (identified by the purple outline) in the central part of Espirito Santo state. Accordingly, we identify this 37 x 37 km area as the highest-priority candidate for new protection, providing some coverage for 98 previously-unprotected species. 8 Figure 4: Brazil – Top 4 candidate areas for new protection 9 Figure 4b reports the counting results when these 98 species are withdrawn from consideration for the second round. Although high counts in the east and southeast are again evident, the maximum cell count in this round is in the northeast part of Amazonas state. This identifies the second-priority area, which provides some coverage for 43 unprotected species. The next iteration identifies the third priority area in the southeast, in eastern Paraná state, which offers some coverage for 32 species. The fourth iteration also identifies the highest-count cell in Paraná, in its northern area, offering coverage for 30 species. Together, these iteration steps identify four 37 km x 37 km areas that provide coverage for 203 (14.4%) of Brazil’s 1,412 unprotected endemic species. Cameroon Figure 6 displays Cameroon’s protected areas in dark green. Its protection intensity is substantially less than Brazil’s, with 55,600 sq km (12%) of Cameroon’s 464,319 sq km in protected areas. In our database Cameroon has 645 endemic species, of which 189 (29.3%) are effectively unprotected. Figure 5 displays images of some unprotected species and Figure 6 displays the ranges and public-domain images of some vertebrates (Figure 6a), plants (6b) and arthropods (6c) from the unprotected group. Their geography has a pattern that resembles Brazil’s: They lie outside existing protected areas (by construction), and their occurrence areas are generally small because larger occurrence areas will often overlap with protected areas. In Cameroon, the displayed species are predominantly located in the country’s west and southwest. 10 Figure 5: Cameroon – Images of unprotected species 11 Figure 6: Cameroon – Images of unprotected species 6a: Vertebrates 6b: Plants 12 6c: Arthropods The four panels of Figure 7 replicate the iterative method that we have introduced in the Brazilian case. Cameroon’s grid scale is 0.136 dd (cells are 15.1 km on a side). Figure 7a displays counts by grid cell for Cameroon’s 189 unprotected endemic species. The figure reveals pronounced clustering in the west and southwest, with the highest cell count (outlined in purple) in the southern part of Centre Region. This cell provides coverage for 24 previously-unprotected species. The three following iterations identify priority areas in the central part of Nord-Ouest Region (21 species, Figure 7b), the southwest corner of Est Region (16 species, Figure 7c) and the southern part of Sud-Ouest Region (15 species, Figure 7d). Together, these iteration steps identify four 15 km x 15 km areas that provide coverage for 76 (40.2%) of Cameroon’s 189 unprotected endemic species. 13 Figure 7: Top 4 priority areas in Cameroon 14 In summary, our introductory case analyses for Brazil and Cameroon have illustrated our basic methodology for identifying unprotected endemic species, establishing grid scales, and selecting high-priority areas for protection. The results suggest that relatively few new protected areas can provide coverage for a substantial portion of unprotected endemic species. However, an important caveat has to be addressed: If the species occurrence maps are accurate, the grid squares identified by our exercise will provide some coverage for all overlapping species. However, our methodology does not guarantee that each species within a grid cell will be protected by our definition. This can be remedied by expanding the protected area until the coverage achieved removes all species from unprotected status as defined by our original criterion (protection for less than 5% of a species’ occurrence map and less than 25 sq km). For Brazil, we compute the needed expansion for the 98 species identified in the first iteration of our area selection methodology. We generate a sequence of circles from the centroid of the identified grid cell, starting with the circle that circumscribes the cell.8 In each iteration, we expand the radius by 10%. We find that the first circle removes 92 of 98 species from unprotected status. Expanding the radius by 10% removes 3 additional species, and the remaining 3 species are removed from unprotected status by respective expansions (from the original radius) of 20%, 40% and 50%. For Cameroon, the first iteration identifies a grid cell that removes 24 species from unprotected status. With the same protection test, the initial circle protects 9 species and successive increases of 20% to 50% of the original radius add protection for 4, 3, 1 and 2 species, respectively. The remaining 5 species require radial expansions of 70%, 80%, 120%, 150% and 160%, respectively. Figure 8 displays the circular areas that provide protection coverage for all species in the priority 1 areas of Brazil and Cameroon. In both cases, these areas fall well within the size range of typical protected areas in the two countries. 8 The radius of the circle is one-half the cell side length. 15 Figure 8: Brazil and Cameroon – Full protection areas for priority group 1 4. Variations in Geography and Taxonomy: South Africa and Costa Rica We turn to South Africa and Costa Rica for a more detailed look at geographic and taxonomic variations that are revealed by our methodology. The two countries differ substantially in size, yielding very different grid scales for priority area identification: 0.183 dd (20.3 km) for South Africa and 0.069 dd (7.7 km) for Costa Rica. South Africa has 12,793 endemic species in our database, with 1,192 unprotected by our criteria. Overall, South Africa’s current protection profile is impressive, with about 91% of its endemic species in protected areas that occupy about 9% of the national territory. Costa Rica’s profile is also impressive, with 97.9% of its endemic species in protected areas that occupy about 59% of the national territory. Of Costa Rica’s 6,075 endemic species in our database, 125 are unprotected by our criteria. South Africa For South Africa, Figure 9 displays images of some unprotected plants, arthropods, vertebrates and other species whose images are in the public domain. Figure 10 displays our identification of the top 3 priority areas for new protection. These results provide a striking picture of geographic variation, with priority area 1 in the eastern coastal region (Figure 10a: KwaZulu – Natal), area 2 in the far west (Figure 10b: Northern Cape) and area 3 in the interior northeast (Figure 10c: Gauteng). At the same time, all three figures show that the priority grid cells have maximum species counts in distributions with relatively high counts across broad areas. 16 Figure 9: South Africa – Selected unprotected species 17 Figure 10: Top 3 priority areas in South Africa Table 2 provides a detailed accounting for the species that would be protected by the three priority areas. Although priority group 1 has the most species by construction, the table shows that all three groups are relatively large. Arthropods dominate group 1, although it also includes substantial numbers of vertebrates, plants and other species. Group 2 has narrower taxonomic representation, confined to plants and arthropods, and the former are much more numerous. Group 3 is also confined to these two groups, but plants are only represented by one species. Table 3 expands the species accounting to the top 10 priority areas, providing information on total species protected and the distribution of species across the four taxonomic groups. The table confirms wide dispersal of unprotected species: 62 species are in priority area 1, 55 are in area 2, and representation tails off gradually to 25 species in area 10. Taxonomic representation exhibits great variation across the 10 areas, with percentage ranges of [2.1 – 96.8] for plants, [0 – 18.5] for vertebrates, [0 – 97.9] for arthropods and [0 – 43.6] for other species. 18 Table 2: South Africa – Species in top 3 protected areas 2a. Group 1 Vertebrates Plants Arthropods Other Species Anchichoerops natalensis Crassula floribunda Aedes demeilloni Achnanthes rupestris Bathygobius niger Eriosema superpositum Aedes hansfordi Bullia mozambicensis Gerrardanthus Canthigaster punctata tomentosus Anoba disjuncta Chalcocystis burnupi Halichoeres zulu Gymnosporia woodii Bandusia rubicunda Diodora crucifera Phymaspermum Brachyplatys Pempheris eatoni pinnatifidum testudonigra Echinolittorina natalensis Thamnaconus arenaceus Senecio exuberans Cabomina hilariformis Gomphonema entolejum Syncolostemon latidens Callioratis abraxas Halgerda dichromis Callyna decora Kerkophorus corneus Carbula recurva Kerkophorus melvilli Cassida guttipennis Medusafissurella dubia Catamonus melancholicus Catephia amplificans Chiasmia natalensis Cicynethus decoratus Drosophila vulcana Glenea arida Helcita wahlbergi Hortipes mesembrinus Hyblaea occidentalium Hypena palpitralis Leptopholcus gracilis Monolepta jacksoni Negera natalensis Phonoctonus fasciatus Plecoptera poderis Plusiopalpa dichora Pseudandriasa mutata Pseudochromatosia nigrapex Raceloma natalensis Rhynocoris venustus Stagira microcephala Syllepte orbiferalis Synema diana Syngamia fervidalis Thysanoplusia spoliata Trichoplusia roseofasciata Veniliodes inflammata Xosophara guttata Zerenopsis geometrina 19 2b. Group 2 and Group 3 Group2 Group 3 Plants Arthropods Plants Arthropods Albuca pentheri Clania macgregori Pearsonia bracteata Ableptina nubifera Antimima gracillima Colletes zygophyllum Acollesis fraudulenta Arctotheca marginata Cryptolarynx estriatus Acontia tanzaniae Babiana papyracea Idopompilus braunsi Acontia tinctilis Chamarea snijmaniae Merostenus longistylus Acontia torrefacta Chrysocoma hantamensis Moluris semiscabra Afrobirthama hobohmi Cyanella aquatica Nesomyrmex nanniae Aphilopota iphia Cygnicollum immersum Nixonia mcgregori Aphilopota subalbata Daubenya capensis Opistophthalmus laticauda Aspilatopsis gloriola Daubenya stylosa Patellapis bifurcata Athetis xanthopis Diascia insignis Patellapis doleritica Audea fumata Diascia lewisiae Patellapis minima Chiasmia kirbyi Euryops mirus Patellapis tenuihirta Clostera lentisignata Geissorhiza divaricata Patellapis timpageleri Concavia transvaali Geissorhiza subrigida Prosoeca marinusi Conchylia sesquifascia Hesperantha rivulicola Scrapter mellonholgeri Cropera sericea Hesperantha vaginata Scrapter oxaloides Desmeocraera vernalis Ixia brunneobractea Drepanogynis glaucichorda Lachenalia alba Epilepia melanobrunnea Massonia pseudoechinata Epiplema inconspicua Mesembryanthemum tenuiflorum Eublemma albivena Moraea aspera Eublemma fulvitermina Moraea fragrans Eutricha fulvida Moraea hesperantha Euxoa anarmodia Moraea pseudospicata Gibbalaria sistrata Oxalis argillacea Halseyia rufilinea Oxalis callosa Idaea associata Oxalis filifoliolata Laelia extatura Pelargonium aristatum Lamoria exiguata Pelargonium pachypodium Leucania cupreata Pentameris dentata Lithacodia normalis Romulea discifera Lymantriades xanthosoma Romulea sabulosa Mauna ava Selago spectabilis Metasia eremialis Strumaria picta Metopteryx rattus Trachyandra prolifera Mimasura innotata Xiphotheca canescens Mimasura quadripuncta Zaluzianskya regalis Mythimna atrinota Naarda melanomma Nycteola malachitis Odontopera homales Omphalucha albosignata Panagropsis muricolor Pseudomallada sjostedti Scopula deserta Scotopteryx cryptocycla 20 21 Table 3: South Africa – Distributions of new priority areas and species % Total Other % % % Other Priority Species Species Plants Vertebrates Arthropods Species Plants %Vertebrates Arthropods Species 1 62 5.2 7 6 39 10 11.3 9.7 62.9 16.1 2 55 4.6 38 0 17 0 69.1 0 30.9 0 3 47 3.9 1 0 46 0 2.1 0 97.9 0 4 40 3.4 38 0 2 0 95 0 5 0 5 39 3.3 20 2 0 17 51.3 5.1 0 43.6 6 31 2.6 30 0 1 0 96.8 0 3.2 0 7 30 2.5 27 1 1 1 90 3.3 3.3 3.3 8 27 2.3 18 5 4 0 66.7 18.5 14.8 0 9 26 2.2 24 1 1 0 92.3 3.8 3.8 0 10 25 2.1 3 1 19 2 12 4 76 8 Costa Rica Arthropods figure prominently in the Costa Rican case, and Figure 11 presents our identification of the top 3 priority areas for new protection. These display more geographic concentration than their South African counterparts, with the top two areas (Figures 11a and 11b) in contiguous subregions of western Alajuela province and the third priority region (Figure 11c) in the central coastal region in Puntarenas Province. In contrast to the South African case, we observe little dispersion of high grid cell counts beyond the priority areas. 22 Figure 11: Top 3 priority areas in Costa Rica 23 Table 4 provides a detailed accounting for the unprotected species that would be protected by the three priority areas in Costa Rica. Table 4: Costa Rica: Species in Top 3 New Protected Areas Group 1 Group 2 Group 3 Arthropods Arthropods Arthropods Other Species Aerophilus sandraberriosae Houghia destituta Apanteles leonelgarayi Aricidea petacalcoensis Alabagrus jennyphillipsae Houghia ochrofemur Euplectrus gerarddelvarei Chaetozone nicoyana Alabagrus scottmilleri Iconella andydeansi Euplectrus mikegatesi Glyphohesione nicoyensis Alabagrus tanyadapkeyae Janhalacaste guanacastensis Exasticolus sigifredomarini Leitoscoloplos multipapillatus Apanteles deifiliadavilae Leurus iangauldi Hormius eddysanchezi Oxydromus minutus Apanteles eulogiosequeirai Macrocentrus harisridhari Lytopylus johanvalerioi Apanteles guadaluperodriguezae Megalota crassana Macrocentrus iangauldi Apanteles hectorsolisi Megalota ochreoapex Snellenius isidrochaconi Apanteles keineraragoni Microplitis francopupulini Apanteles luisbrizuelai Neurophyseta completalis Apanteles marisolarroyoae Niconiades gladys Apanteles monicachavarriae Ophionellus albofasciatus Apanteles ronaldquirosi Orgilus ebbenielsoni Bracon andrewwalshi Orgilus elizabethpennisiae Cardiochiles alejandrosolorzanoi Orgilus johnpipolyi Cardiochiles rogerblancoi Parapilocrocis albomarginalis Catharylla myrabonillae Plesiocoelus vanachterbergi Chelonus hartmanguidoi Potamanaxas effusa Chelonus jimlewisi Prasmodon johnbrowni Chelonus nataliaivanovae Prasmodon scottmilleri Chelonus normwoodleyi Pseudapanteles carlosespinachi Chelonus robertoespinozai Sparganocosma docsturnerorum Chelonus scottmilleri Spathidexia aurantiaca Chlamydastis irenecanasae Stantonia henrikekmani Cordyligaster fuscipennis Stenoma leucana Cosmorrhyncha albistrigulana Telothyria eldaararyae Cubus montywoodi Telothyria incisa Cystomastax angelagonzalezae Telothyria manuelpereirai Dolichogenidea josealfredohernandezi Telothyria ricardocalerloi Dynamine gillotti Trichomma batistai Enosis pruinosa Trigonospila uniformis Eriglenum tristum Wilkinsonellus alexsmithi Euplectrus eowilsoni Xiphosomella belinda Euplectrus victoriapookae Yelicones dirksteinkei Gonioterma ignobilis Gonodonta pulverea Himmacia inflammata Hormius jorgecarvajali Hormius manuelzumbadoi 24 Here the pattern of geographic clustering is joined to a pattern of extreme clustering by species group. The species in priority group 1 are dominant numerically, and all are arthropods. All species in group 2 are arthropods as well. Group exclusivity persists in group 3, although it is populated entirely by other species. Table 5 provides an expanded view for the top 10 priority areas, and the count pattern across priority groups provides an extreme contrast to the South African case. In Costa Rica, the area 1 group of 73 species declines sharply to 8 species in group 2, 5 in group 3, and small numbers for groups 4 – 10. Arthropods are overwhelmingly dominant in the top groups, while the smaller groups have varying patterns of representation for plants, vertebrates, arthropods and other species. Table 5: Costa Rica – Distributions of new priority areas and species % Total Other % % % Other Priority Species Species Plants Vertebrates Arthropods Species Plants %Vertebrates Arthropods Species 1 73 58.4 0 0 73 0 0 0 100 0 2 8 6.4 0 0 8 0 0 0 100 0 3 5 4 0 0 0 5 0 0 0 100 4 4 3.2 0 0 4 0 0 0 100 0 5 3 2.4 2 0 0 1 66.7 0 0 33.3 6 3 2.4 0 1 1 1 0 33.3 33.3 33.3 7 3 2.4 0 0 3 0 0 0 100 0 8 2 1.6 1 0 0 1 50 0 0 50 9 2 1.6 0 0 0 2 0 0 0 100 10 2 1.6 0 0 2 0 0 0 100 0 In summary, the South African and Costa Rican cases illustrate the potential diversity of outcomes when our methodology for identifying priority areas is applied to countries with very different patterns of geographic dispersion and taxonomic representation for unprotected species. Historically, vertebrates have commanded much of the attention when protected areas have been delineated. However, these two country cases suggest much more important roles for other taxonomic groups in the identification of priority areas for future protection. In addition, the contrasting rates of decline in priority area species counts suggest significant variations in the land opportunity costs of movement toward greater species protection. Opportunity costs may be significantly lower in countries like Costa Rica, where the remaining unprotected species are highly clustered in a few areas. In countries like South Africa, on the other hand, widely-dispersed clusters of unprotected species will require more extensive land acquisition to meet ambitious conservation goals. 25 5. Spatial Implications of Full Species Protection: Ecuador and Papua New Guinea In this section, we explore the implications of movement toward full protection of endemic species with illustrations from Ecuador and Papua New Guinea. The two countries are in the same size class, so they have similar grid scales for priority identification: 0.113 dd (12.5 km) for Ecuador and 0.136 dd (15.1 km) for Papua New Guinea. Ecuador has 3,477 endemic species in our database, with 698 species that are unprotected by our criteria. Ecuador’s protected areas occupy about 23% of the national territory, offering protection for about 80% of its endemic species. In contrast, Papua New Guinea’s protected areas occupy about 4% of the national territory, affording protection to 61% of its 3,266 endemic species in our database. In Papua New Guinea, 1,270 endemic species remain unprotected by our criteria. For this exercise, we run our algorithm for priority area identification until all unprotected species have coverage in new protected areas. We divide priority areas into decile groups based on cumulative species protected. Figure 12 displays the geographic implications for Ecuador. The priority areas are grid cells with scales identified above, but we size the colored symbols for both visibility and reflection of relative importance. As the first panel shows, 20% of unprotected species are afforded coverage by two new priority areas (purple) at opposite ends of the country, in Pichincha and Loja Provinces. The next 10% are covered by two areas in the south, in Azuay and Zamora Chinchipe Provinces. In a similar vein, relatively few priority areas cover the unprotected species in the third, fourth and fifth groups. After that, the number of species covered by new areas begins to decline steadily and the number of areas grows proportionately. A very large number of new areas are required for the final species decile. They are in almost every province and particularly numerous in the axis from Esmeraldas and Carchi Provinces in the north to Loja and Zamora Chinchipe Provinces in the south. Figure 13 displays analogous results for Papua New Guinea, and a similar pattern emerges as we move through the deciles. In the top four priority classes, the number of identified areas grows steadily: 1 (Madang Province); 2 (Madang, Oro); 3 (Western Highlands, Morobe, Milne Bay); 4 (Sandaun, Gulf, Milne Bay, New Ireland (2 areas); 5 (Western, Chimbu, Morobe, Central, Milne Bay, Manus). Then the required number of areas begins expanding as the number of covered species per area declines. In the lowest-priority panel, the country is covered with a dense pattern of dots that are present in every province. 26 Figure 12: Ecuador - Incremental protected areas by new species coverage percent decile 27 Figure 13: Papua New Guinea - Incremental protected areas by new species coverage percent decile 28 Figures 14 and 15 explore the territorial implications for the two countries using the data generated by iterative identification of new protected areas and expansion of protected areas to ensure protection for all species. In Ecuador (Figure 14), the iterative process begins with 23% of national territory already under protection. The horizontal axis measures the percent of unprotected species that are cumulatively added to new protected areas. As the graph shows, covering 50% of unprotected species requires an additional 4% of national territory (to a total of 27%). Then the curve begins to steepen, with about 34% of national territory required for 80% unprotected species coverage, 44% for 95% coverage, and 48% for 100% coverage. In Papua New Guinea (Figure 15), the iterative process begins with only 4% of national territory already under protection. As the graph shows, covering 50% of unprotected species requires 6%. Then the curve steepens, with about 11% of national territory required for 80% unprotected species coverage, 18% for 95% coverage, and 24% for 100% coverage. Figure 14: Ecuador - Species protection vs. land in protected areas 29 Figure 15: Papua New Guinea - Species protection vs. land in protected areas These results for Ecuador and Papua New Guinea suggest that replication of our methodology for every country could yield valuable insights for the global 30x30 initiative. Moving toward 100% coverage for unprotected endemic species would move countries like Ecuador beyond the 30% territorial coverage target, while other countries like Papua New Guinea would remain below it. For each country, the implications would depend on the percentage of national territory that is already protected, the percentage of endemic species covered by existing protection, and the geographic distribution of unprotected species in different taxa. 6. Marine Species Protection – Philippines and Madagascar The 30x30 initiative also incorporates protection for marine species. We address the marine protection problem with the same methods that we have applied to terrestrial protection. The only major difference is the venue, since a definition of national marine territory is required. We employ Extended Economic Zones (EEZs), the areas in which countries have exclusive rights to the use of marine resources. The outer boundary of a country’s EEZ is 200 nautical miles (370.4 km) from its coast. In this context, we identify a marine species as endemic to a country if its occurrence area lies within the boundary of the country’s EEZ. As before, we identify a species as effectively 30 unprotected if its overlap with protected areas is less than 25 sq km and less than 5% of its occurrence area. We change our method for setting grid cell resolutions because, in contrast with the terrestrial case, marine protected area sizes vary enormously across countries whose EEZ areas are comparable. For this exercise, we set each country’s grid scale to create 2,500 grid cells within its EEZ. For small-EEZ cases, we generate fewer than 2,500 grid cells by setting a lower bound of 1 km for cell side length. In this section, we explore marine protection for Philippines and Madagascar. Our measure of Philippines’ EEZ area is 1,964,764 sq. km., of which 32,053 sq km (1.6%) are within 103 marine protected areas. Philippines has 82 endemic marine species in our database, of which 55 (67%) are protected by our criteria. Madagascar’s EEZ area is 1,196,285 sq km, of which 14,457 (1.2%) lie within 79 protected areas. Madagascar has 67 marine endemic species, of which 43 (64.2%) are protected. Philippines Philippines’ grid resolution is 0.253 dd (28.1 km). We replicate our terrestrial methodology for the marine case, and Figure 16 identifies the four highest-priority sites for marine protection. The highest-priority area, covering 8 unprotected species, lies off of the northwest coast directly west of Batangas Province. The second-priority area, covering 7 species, lies off the northern coast of Palawan Province. Area 3, covering 3 species, lies off the northeast coast directly offshore from Aurora Province, and area 4, covering 2 species, lies off the northern coast of the main island of Palawan Province. When compared with the terrestrial cases, our results suggest that moving to full species protection in Philippines would impose very modest requirements on its EEZ. As Figure 17 shows, coverage of all unprotected species would move the current protected marine area from 1.7% to 2.3%. 31 Figure 16: Top 4 marine priority areas in Philippines 32 Figure 17: Philippines - Marine species protection vs. territory in EEZ Madagascar Madagascar’s grid resolution is 0.197 dd (21.9 km). Figure 18 identifies the four highest-priority sites for marine protection. The highest-priority area, covering 17 unprotected species, lies off Androy Region in the southernmost part of the country. The second-priority area, covering 2 species, lies off the neighboring region of Anosy. Area 3, covering 1 species, lies off the northern end of the Vatovavy Fitovinany Region on the country’s east coast. Area 4, covering 1 species, lies off of Sofia Region in the north. As in the Philippines case, our results (Figure 19) suggest that moving to full species protection in Madagascar would impose very modest requirements on its EEZ. Coverage of all unprotected species would require only about 1.65% of the country’s EEZ area. 33 Figure 18: Top 4 marine priority areas in Madagascar’s EEZ 34 Figure 19: Madagascar - Marine species protection vs. territory in EEZ 7. National Templates: India and China India and China are among the most important countries for global biodiversity. India has 2,635 endemic species in our database and China has 10,367. Their participation in the global 30x30 initiative will be critical for its success, and we believe that it is important to provide a view of their prospects for protecting the diverse set of new species that are mapped in our database. However, our case studies for these two countries must differ from the previous eight because information about their current species protection status is more limited. India has 17 protected areas in the World Database of Protected Areas, but the WDPA notes that it has 900 protected areas which are not publicly reported.9 China has 87 protected areas in the WDPA, but 2,960 are 9 Citation: Protected Planet (2024). “India chooses to restrict some data on its protected areas. As a result, data on 900 protected areas is not publicly-available and cannot be viewed or downloaded on this page. While these sites are included in the coverage statistics presented here, all other statistics (e.g. number of protected areas; breakdown by governance type etc.) on this page are based upon the publicly- available data only.” https://www.protectedplanet.net/country/IND 35 not publicly reported.10 By implication, our estimates of 1,211 unprotected species for India and 2,156 for China must be significantly higher than the actual numbers. Nevertheless, we believe that analyses for the two countries are worthwhile because our methodology can provide comparative templates that may be useful for colleagues in India and China whose information is more complete than ours. In particular, the inclusion of many new species in our GBIF database may provide new insights about priority areas for protection in the two countries. India Our Indian case methodology goes through 175 iterations to identify candidate areas for protecting all of the 1,211 species that are not covered by publicly-reported protected areas. Table 6 displays results for the first 40 iterations, which account for 958 (79.1%) of the unprotected species. The distribution is very skewed, with the first 3 areas accounting for nearly 25% of unprotected species and 11 areas accounting for 50%. Within the new protected areas, the diversity of species representation is striking. Representation varies from 0 to 87.5% for plants, 0 to 82.9% for vertebrates, 0 to 100% for arthropods and 0 to 100% for other species. 10 Citation: Protected Planet (2024). “China chooses to restrict some data on its protected areas. As a result, data on 2,960 protected areas is not publicly-available and cannot be viewed or downloaded on this page. While these sites are included in the coverage statistics presented here, all other statistics (e.g. number of protected areas; breakdown by governance type etc.) on this page are based upon the publicly- available data only.” https://www.protectedplanet.net/country/CHN 36 Table 6: India – Distributions of new priority areas and species % Other Priority Species Species % Cum % % Plants %Vertebrates % Arthropods Species 1 117 9.7 9.7 24.8 10.3 62.4 2.6 2 93 7.7 17.3 8.6 3.2 28.0 60.2 3 86 7.1 24.4 61.6 11.6 24.4 2.3 4 46 3.8 28.2 4.3 10.9 52.2 32.6 5 41 3.4 31.6 68.3 7.3 14.6 9.8 6 41 3.4 35.0 4.9 4.9 90.2 0.0 7 41 3.4 38.4 7.3 82.9 7.3 2.4 8 37 3.1 41.5 0.0 0.0 100.0 0.0 9 36 3.0 44.4 41.7 16.7 36.1 5.6 10 35 2.9 47.3 28.6 0.0 40.0 31.4 11 31 2.6 49.9 29.0 45.2 19.4 6.5 12 24 2.0 51.9 29.2 16.7 45.8 8.3 13 23 1.9 53.8 56.5 8.7 17.4 17.4 14 20 1.7 55.4 40.0 20.0 35.0 5.0 15 19 1.6 57.0 31.6 21.1 31.6 15.8 16 19 1.6 58.5 10.5 36.8 42.1 10.5 17 17 1.4 60.0 11.8 0.0 76.5 11.8 18 16 1.3 61.3 0.0 25.0 75.0 0.0 19 16 1.3 62.6 0.0 75.0 25.0 0.0 20 16 1.3 63.9 31.2 25.0 31.2 12.5 21 15 1.2 65.2 0.0 0.0 0.0 100.0 22 14 1.2 66.3 50.0 7.1 14.3 28.6 23 12 1.0 67.3 0.0 8.3 83.3 8.3 24 12 1.0 68.3 41.7 33.3 16.7 8.3 25 12 1.0 69.3 16.7 25.0 41.7 16.7 26 12 1.0 70.3 33.3 16.7 41.7 8.3 27 11 0.9 71.2 63.6 27.3 9.1 0.0 28 9 0.7 71.9 77.8 0.0 11.1 11.1 29 9 0.7 72.7 0.0 22.2 11.1 66.7 30 9 0.7 73.4 77.8 0.0 0.0 22.2 31 8 0.7 74.1 37.5 0.0 62.5 0.0 32 8 0.7 74.7 87.5 0.0 12.5 0.0 33 8 0.7 75.4 50.0 25.0 25.0 0.0 34 8 0.7 76.1 0.0 25.0 50.0 25.0 35 7 0.6 76.6 57.1 14.3 14.3 14.3 36 7 0.6 77.2 28.6 0.0 71.4 0.0 37 7 0.6 77.8 14.3 0.0 71.4 14.3 38 6 0.5 78.3 16.7 0.0 83.3 0.0 39 5 0.4 78.7 20.0 0.0 80.0 0.0 40 5 0.4 79.1 0.0 40.0 40.0 20.0 Figure XX: India: Species Protection vs. Land in Protected Areas 37 We present the complete species allocation results for 175 iterations in Figure 20. New protected areas are in grid cells that surround the centroids of circles that are sized and color-coded by number of new species protected (from purple through red to orange and yellow). The three largest (purple) circles identify areas in Maharashtra, Telangana and Karnataka that provide coverage for about 25% of all unprotected species. The next-priority (dark red) circles in Gujarat, Tamil Nadu, Uttarakhand and Andaman and Nicobar account for another 14%, bringing cumulative coverage to 38%. The next-priority (red) circles in Kerala, southern and western Maharashtra and Sikkim add 12% and bring cumulative coverage to 50%. The remaining 50% are protected in areas that are scattered across the country, with particularly large concentrations in the northern mountain region and the Western Ghats. As Figure 21 shows, India’s spatial requirements for complete coverage of the unprotected species are relatively modest by the 30x30 standard. These calculations begin with India’s publicly-reported protected areas, which occupy less than 1% of the national territory. As the figure shows, covering 50% of the unprotected species in our analysis requires only 1.3% of India’s territory. This rises to 2.6% for 80% coverage, 5.4% for 95% coverage, and 7.7% for 100% coverage. 38 Figure 20a: Northern India – New protected areas for complete species coverage 39 Figure 20b: Eastern India – New protected areas for complete species coverage 40 Figure 20c: Southern India – New protected areas for complete species coverage 41 Figure 21: India: Species protection vs. land in protected areas China For China, our methodology goes through 303 iterations to identify candidate areas for protecting all of the 2,156 species that are not covered by publicly-reported protected areas. Table 7 displays results for the first 40 iterations, which account for 1,330 (61.7%) of the unprotected species. China’s distribution is significantly less skewed than India’s, requiring 7 areas to account for 25% of the unprotected species and 23 areas to account for 50%. Again, the diversity of species representation within new protected areas is very high. Representation varies from 8.3% to 100% for plants, 0 to 25.4% for vertebrates, 0 to 81.2% for arthropods and 0 to 83.3% for other species. 42 Table 7: China – Distributions of new priority areas and species % Other Priority Species Species % Cum % % Plants %Vertebrates % Arthropods Species 1 103 4.8 4.8 24.3 15.5 54.4 5.8 2 101 4.7 9.5 9.9 1.0 81.2 7.9 3 96 4.5 13.9 30.2 3.1 61.5 5.2 4 73 3.4 17.3 75.3 1.4 12.3 11.0 5 71 3.3 20.6 38.0 18.3 32.4 11.3 6 62 2.9 23.5 93.5 0.0 6.5 0.0 7 59 2.7 26.2 49.2 25.4 25.4 0.0 8 58 2.7 28.9 55.2 0.0 43.1 1.7 9 52 2.4 31.3 96.2 0.0 3.8 0.0 10 47 2.2 33.5 78.7 2.1 19.1 0.0 11 42 1.9 35.4 40.5 16.7 40.5 2.4 12 37 1.7 37.2 13.5 0.0 78.4 8.1 13 33 1.5 38.7 90.9 6.1 3.0 0.0 14 33 1.5 40.2 78.8 6.1 12.1 3.0 15 32 1.5 41.7 37.5 6.2 46.9 9.4 16 31 1.4 43.1 90.3 3.2 6.5 0.0 17 10 0.5 43.6 70.0 10.0 10.0 10.0 18 25 1.2 44.8 92.0 4.0 4.0 0.0 19 24 1.1 45.9 83.3 0.0 12.5 4.2 20 23 1.1 46.9 87.0 4.3 0.0 8.7 21 21 1.0 47.9 19.0 9.5 71.4 0.0 22 20 0.9 48.8 100.0 0.0 0.0 0.0 23 20 0.9 49.8 95.0 0.0 5.0 0.0 24 19 0.9 50.6 36.8 10.5 47.4 5.3 25 19 0.9 51.5 68.4 10.5 21.1 0.0 26 18 0.8 52.4 50.0 0.0 44.4 5.6 27 18 0.8 53.2 83.3 0.0 11.1 5.6 28 17 0.8 54.0 94.1 0.0 5.9 0.0 29 16 0.7 54.7 62.5 6.2 31.2 0.0 30 16 0.7 55.5 100.0 0.0 0.0 0.0 31 15 0.7 56.2 100.0 0.0 0.0 0.0 32 15 0.7 56.9 93.3 6.7 0.0 0.0 33 14 0.6 57.5 42.9 0.0 50.0 7.1 34 14 0.6 58.2 100.0 0.0 0.0 0.0 35 13 0.6 58.8 76.9 7.7 7.7 7.7 36 13 0.6 59.4 53.8 7.7 38.5 0.0 37 13 0.6 60.0 92.3 0.0 7.7 0.0 38 13 0.6 60.6 69.2 15.4 15.4 0.0 39 12 0.6 61.1 8.3 0.0 8.3 83.3 40 12 0.6 61.7 83.3 8.3 8.3 0.0 43 We present the complete species allocation results for 303 iterations in Figure 22, with circles again sized and color-coded by number of new species protected. The three largest (purple) circles identify areas in Hainan, Jiangsu and Beijing that provide coverage for 14% of all unprotected species. The next-priority (dark red) circles in Yunnan, Sichuan, Shaanxi and Fujian account for another 15%, bringing cumulative coverage to 29%. The next-priority (red) circles in Xizang, Sichuan and Guangdong bring cumulative coverage to 35%. The remaining 65% are protected in areas that are scattered across the country, with particularly large concentrations in southwest China. 44 Figure 22: China: New protected areas for complete species coverage 45 As Figure 23 shows, the spatial requirements for complete coverage of unprotected species are somewhat greater than India’s (reflecting the greater spatial dispersion in China), but still relatively modest by the 30x30 standard. Our calculations begin with China’s publicly-reported protected areas, which occupy 1.7% of the national territory. As the figure shows, covering 50% of the unprotected species in our analysis requires 2.3% of China’s territory. This rises to 3.9% for 80% coverage, 6.7% for 95% coverage, and 9.3% for 100% coverage. Figure 23: China - Species protection vs. land in protected areas 8. Summary and Conclusions The publication of nearly 600,000 new species occurrence maps from GBIF data by Dasgupta et al. (2024) has provided an opportunity to revisit international species protection with much broader representation for plants, invertebrates and other species. This is a particularly opportune moment, because the global 30x30 initiative includes commitments from 188 governments to expand terrestrial and marine protection toward 30% of the globe by 2030. This paper has aimed to contribute by mobilizing the GBIF occurrence maps to explore new opportunities for species protection in a sample of 10 countries in Latin America (Brazil, Costa Rica, Ecuador), Africa (Cameroon, South Africa, Madagascar) and the Asia-Pacific region (Papua 46 New Guinea, Philippines, India and China). We focus on individual countries to highlight the role of conservation stewardship in local settings. We consider both terrestrial and marine cases, focusing on species that are endemic to each country. Our approach departs from many previous exercises by giving equal weight to all vertebrates, invertebrates, plants, and other species whose occurrence regions are mapped in our database. We use a spatially-efficient algorithm to identify a hierarchy of priorities for localities where new protected areas would offer coverage to unprotected species. Our 10 country cases draw on our results to explore the spatial implications for countries’ 30x30 commitments. We find that initial conditions matter greatly, because some countries already protect large parts of their territories while others do not; unprotected species are spatially clustered in some countries but not others; and the representation of different taxa among unprotected species varies greatly across countries. In consequence, full protection for newly- mapped species can be achieved inside the 30% territorial standard for some countries, while others would have to go well beyond that limit to achieve the same result. In all cases, however, our results show that spatial clustering is sufficient to yield very significant protection gains for relatively modest expansion of protected areas. In summary, we believe that our methodology has yielded sufficient insight to warrant expansion to a much larger group of countries. We have treated all taxa as equal for these cases, but we recognize that not all countries may choose this approach. Similarly, our unique focus on endemics and our criteria for identifying unprotected species may not suit the preferences of environmental policy makers in all countries. At the same time, we believe that our iterative approach to the identification of high-priority areas for new protection can be useful, whatever species weighting criteria are employed. 47 References Andam, K. S., Ferraro, P. J., Pfaff, A., Sanchez-Azofeifa, G. A., & Robalino, J. A. (2008). Measuring the effectiveness of protected area networks in reducing deforestation. Proceedings of the national academy of sciences, 105(42), 16089- 16094. Bruner, A. G., Gullison, R. E., Rice, R. E., & Da Fonseca, G. A. (2001). Effectiveness of parks in protecting tropical biodiversity. science, 291(5501), 125-128. Chen, H., Zhang, T., Costanza, R., & Kubiszewski, I. (2023). Review of the approaches for assessing protected areas' effectiveness. Environmental Impact Assessment Review, 98, 106929. Geldmann, J., Barnes, M., Coad, L., Craigie, I. D., Hockings, M., & Burgess, N. D. (2013). Effectiveness of terrestrial protected areas in reducing habitat loss and population declines. Biological Conservation, 161, 230-238. Ghosh-Harihar, M., An, R., Athreya, R., Borthakur, U., Chanchani, P., Chetry, D., ... & Price, T. D. (2019). Protected areas and biodiversity conservation in India. Biological Conservation, 237, 114-124. Joppa, L., & Pfaff, A. (2010). Reassessing the forest impacts of protection: the challenge of nonrandom location and a corrective method. Annals of the New York Academy of Sciences, 1185(1), 135-149. Leverington, F., Costa, K. L., Pavese, H., Lisle, A., & Hockings, M. (2010). A global analysis of protected area management effectiveness. Environmental management, 46, 685-698. Llorente-Culebras, S., Ladle, R. J., & Santos, A. M. (2023). Publication trends in global biodiversity research on protected areas. Biological Conservation, 281, 109988. Ma, B., Zhang, Y., Hou, Y., & Wen, Y. (2020). Do protected areas matter? A systematic review of the social and ecological impacts of the establishment of protected areas. International journal of environmental research and public health, 17(19), 7259. Mora, C., & Sale, P. F. (2011). Ongoing global biodiversity loss and the need to move beyond protected areas: a review of the technical and practical shortcomings of protected areas on land and sea. Marine ecology progress series, 434, 251-266. Nelson, A., & Chomitz, K. M. (2011). Effectiveness of strict vs. multiple use protected areas in reducing tropical forest fires: a global analysis using matching methods. PloS one, 6(8), e22722. Pulido-Chadid, K., Virtanen, E., & Geldmann, J. (2023). How effective are protected areas for reducing threats to biodiversity? A systematic review protocol. Environmental Evidence, 12(1), 18. Rodríguez-Rodríguez, D., & Martínez-Vega, J. (2022). Effectiveness of Protected Areas in Conserving Biodiversity. A Worldwide Review. Cham, Switzerland: Springer. 48 Singh, M., Griaud, C., & Collins, C. M. (2021). An evaluation of the effectiveness of protected areas in Thailand. Ecological Indicators, 125, 107536. Vimal, R., Navarro, L. M., Jones, Y., Wolf, F., Le Moguédec, G., & Réjou-Méchain, M. (2021). The global distribution of protected areas management strategies and their complementarity for biodiversity conservation. Biological Conservation, 256, 109014. Watson, J. E., Dudley, N., Segan, D. B., & Hockings, M. (2014). The performance and potential of protected areas. Nature, 515(7525), 67-73 49 50