Subnational biodiversity reporting metrics for mountain ecosystems

Biodiversity indicators are used to assess progress towards conservation and sustainability goals. However, the spatial scales, methods and assumptions of the underlying reporting metrics can affect the provided information. Using mountain ecosystems as an example, we compare biodiversity protection at subnational scale using the site-based approach of the 2030 Agenda for Sustainable Development (SDG indicator 15.4.1) with an area-based approach compatible with the targets of the Kunming– Montreal Global Biodiversity Framework.

Biodiversity indicators are used to assess progress towards conservation and sustainability goals.However, the spatial scales, methods and assumptions of the underlying reporting metrics can affect the provided information.Using mountain ecosystems as an example, we compare biodiversity protection at subnational scale using the site-based approach of the 2030 Agenda for Sustainable Development (SDG indicator 15.4.1) with an area-based approach compatible with the targets of the Kunming-Montreal Global Biodiversity Framework.
Sustainable development and nature conservation take place in heterogeneous socioeconomic, environmental and policy contexts characterized by unevenly distributed biodiversity 1 .Hence, conservation-related indicator maps calculated at the level of entire countries provide high-level overviews critical for raising awareness but often provide little information about the management actions taken or needed at relevant scales 2,3 .Overcoming these limitations requires subnational maps based on steadily improving spatial data layers that account for the uneven distribution of biodiversity within countries.Such maps enable the assignment of responsibilities between parties and among conservation actors, reveal the need for complementarity in governance and action within and across state boundaries, and help ensure progress towards conservation and the Sustainable Development Goals (SDGs) while 'leaving no one behind', a core value and promise of the United Nations (UN) 2030 Agenda for Sustainable Development 4 .However, although available at increasing resolution, spatial data layers are not always used for breaking down indicators from the national scale to the scale at which policymaking, planning and conservation happen.Mountains are a quintessential example in this context.They host exceptionally rich and functionally important biodiversity 5,6 ; differ in their species' diversity, spatial distribution and levels of endemism across latitudinal, longitudinal and elevational gradients [5][6][7] ; and represent distinct social-ecological systems and landscape units 8 that fall under different jurisdictions within and between countries.As such, they constitute pertinent conservation units and are acknowledged as a conservation priority 9 in the face of accelerating global change.Yet reporting on mountain biodiversity conservation in the context of the 2030 Agenda for Sustainable Development (SDG indicator 15.4.1, coverage by protected areas of important sites for mountain biodiversity 10 ) is performed only at the scale of entire nations, regions or the world.This is the case despite the existence of mountain delineations that enable reporting on biodiversity protection at the level of mountain ranges and systems 8 .In this Brief Communication, we address this gap by applying our most recent mountain inventory 8 to generate spatially disaggregated annual maps of SDG indicator 15.4.1.By doing so, we use the mountain context as an example to illustrate the feasibility of spatially explicit reporting.Moreover, the reporting by means of SDG indicator 15.4.1 does not account for the extent of the Key Biodiversity Areas (KBAs, 'sites that contribute significantly to the global persistence of biodiversity' 11 ) under protection, whereas the action-oriented global targets adopted under the Kunming-Montreal Global Biodiversity Framework 12 could be interpreted in terms of the total percentage of areas under protection.We therefore compare averaged site-level percentages protection (the currently used 'site-based approach') with area-based values that we calculate as the ratios between the protected fractions and the total extent of mountain KBAs ('area-based approach'), highlighting differences and complementarities.To enable the use of our findings by different scientific, management and policy groups, we maintain a web platform to explore and compare spatio-temporal trends in the indicator https://doi.org/10.1038/s41893-023-01232-3and indicate where-and by how much-protection needs to increase.Our approach thus enables annual reporting at spatial scales relevant for policymaking, prioritization and management and contributes to improving the coherence of environmental policies for mountains across scales.
The disaggregation further reveals differences in how well countries protect their shares of transboundary mountain systems.In the European Alps, for example (Fig. 2), levels of protection are lower in Switzerland (~30%) than in neighbouring countries such as Italy (~70%) and Germany (>95%).Given that mountain ecosystems, species and most environmental threats do not stop at political boundaries, such at different scales and according to different calculations (https:// mountainbiodiversity.net/).Further, we provide one-page summaries for countries and mountain systems (Fig. 1; ref. 13) and the R code to replicate the results at mountain range level 14 as open access material.
The disaggregation of SDG indicator 15.4.1 (Fig. 1a) demonstrates that with protection levels of KBAs varying from 0 to 100% across mountain ranges within countries (for example, Bhutan or Switzerland; Supplementary Table 1 and Fig. 1a (2) and ( 4)), the information content of country-level indicators is limited.Disaggregated calculations reveal the spatial variability in biodiversity protection within and across countries that is concealed in unique country-level averages 15 Belize: 88.8 (2) (2) * Switzerland: 21.3 (2) -Sierra Leone: 79.4 (2) (3) https://doi.org/10.1038/s41893-023-01232-3differences might compromise conservation efforts, in particular when countries have very unequal shares of mountain systems.Yet overall, transboundary protection objectives that do not depend on equally high levels of protection across involved countries might nevertheless be achieved.Our results therefore enable the informed assignment of responsibilities between parties and support the implementation of transboundary conservation as a process of international cooperation 16,17 to overcome the social, economic and environmental challenges faced by mountain biodiversity and its beneficiaries.
Comparing the site-and area-based approaches for calculating the levels of protection of mountain KBAs (Supplementary Table 1 and Extended Data Figs. 2 and 3) reveals differences between national-level SDG indicator 15.4.1 values (Fig. 1b).This is the case for nearly 90% of the countries (Supplementary Table 2), with an average difference of 13 percentage points.In more than 65% of these 190 countries, area-based indicator values are higher than site-based values.Within individual countries, differences in SDG indicator 15.4.1 values can be considerable.This is the case for the mountainous country of Bhutan, for example, with 35 percentage points in 2020 (47% and 82%, respectively).This highlights the importance and consequences of methodological choices and calls for caution in the choice of reporting metrics and for careful interpretation of computed values.
The current site-based approach calculates the average percentage coverage by PAs and other effective area-based conservation measures (OECMs) of all mountainous KBAs within a country 10 .The 'mountainous' classification is based on a minimum overlap of 5% between a KBA and terrain that is considered in mountains according to the mountain definition of the UN Environment Programme World Conservation Monitoring Centre (UNEP-WCMC) 18 , which, as a result of its calculation, includes comparatively larger areas of flat and rolling  https://doi.org/10.1038/s41893-023-01232-3terrain.This implies that mountainous KBAs can consist of sizeable portions of hills and lowlands, especially for very large KBAs.Further, according to the site-based calculation, PAs are intersected with entire KBAs and not only with their strictly mountainous fractions.Accordingly, the percentage PA for any given KBA is attributed to mountains even when the PA covers the lowland parts of the mountain KBA.Finally, the treatment of all KBAs as equal regardless of their size (Fig. 1d) helps account for the fact that smaller sites may be particularly effective for conservation, especially for range-restricted species 19 .However, this can also lead to inflated indicator values (Fig. 1d, scenario 1), in particular if the size distribution of KBAs is skewed towards smaller sites (Extended Data Figs. 2 and 3).
The area-based calculation sums the actual area of mountain terrain (sensu ref. 8) that intersects with KBAs and calculates what percentage of this terrain is protected.Accordingly, indicator values increase only with increasing area of protected mountain KBAs.By using a more conservative mountain definition that mostly excludes less-rugged hills and lowlands, and by defining mountain KBAs as the actual area of land that is both mountainous and important for biodiversity, the area-based method overcomes the issue of reporting on PA coverage of land that is not in fact in mountains.Area-based values thereby reflect efforts to protect areas of importance for the persistence of biodiversity that are both large and located strictly in mountains.However, in the presence of at least one large site benefiting from a high protection, overall values are likely to be high despite the possible existence of poorly protected smaller sites (Fig. 1d, scenario 2).
Currently, SDG indicator 15.4.1 quantifies the average percentage PA coverage of sites that are important for mountain biodiversity persistence and located (partly) in mountains 10 .The area-based calculation of SDG indicator 15.4.1 at the subnational level sheds a different and complementary light on mountain biodiversity protection and highlights areas within mountain regions in need of subnational to international conservation efforts.Disaggregation and area-based calculations do not resolve the known shortcomings of KBAs for reporting on biodiversity conservation 19,20 .However, they enable the quantification of progress towards the area-based action-oriented targets of the Kunming-Montreal Global Biodiversity Framework.Moreover, the methodological considerations also apply for other (non-KBA) designations of importance.
Our work on SDG indicator 15.4.1 highlights the need for transparency and caution with regard to the methods and assumptions underlying metrics for informing on sustainability and conservation.With our online resources, we offer tools to support science-based decision-making for complex mountain environments.We enable users to perform their own calculations and comparisons at the relevant scale, understand what indicators mean and conceal, compare different methods and assumptions, and explore time series across mountains.Our online resources support users across sectors and institutions in correctly interpreting the metrics on which decisions and management are based.
Lowland Mountain not KBA not protected Protected mountain not KBA Protected mountain KBA Mountain KBA not protected Bhutan: 81.7

Fig. 1 |
Fig. 1 | KBA coverage by PA. a, Area-based SDG indicator 15.4.1 for mountain ranges of Switzerland, Bhutan, Belize and Sierra Leone: (1) relative country size compared with the smallest; (2) bivariate map of KBA coverage by PA; (3) country area partitioning into lowland and mountains (1 square approximates 1%); (4) number of ranges per subnational indicator values (excludes ranges with no KBA).b, Comparison between area-and site-based calculations.Red symbols, countries in a; circles, other countries; diagonal, same value for both calculations (not considering a country's mountain area).c, (1) legend for a (2 and 4); (2) legend for a (3).d, Comparison between area-and site-based calculations for three scenarios of PA coverage of two KBAs differing in size (green, protected KBA; red, KBA not protected).For illustration purposes, site-based values are identical despite differences in the extent of protected KBA (Extended Data Fig. 2).

Fig. 2 |
Fig. 2 | Transboundary protection of mountain biodiversity.a, KBA coverage by PA across transboundary ranges (dotted areas).Light green, PA not KBA; dark green, protected KBA; red, KBA not protected.Right: site-/area-based national value for Alpine countries neighbouring Switzerland.b,c, Bivariate map of KBA coverage by PA, disaggregated to the mountain range level (b) by country (c).