Climate change and biodiversity loss are major challenges. Habitat loss and degradation, exacerbated by climate change, threaten biodiversity and ecosystem services. Protected areas (PAs) are crucial for conservation, and expanding the PA network is vital for maintaining intact ecosystems and achieving natural climate solutions. Growing calls advocate for ecosystem-level targets for PA expansion, using ecoregions as surrogates for biodiversity. Proposals, such as the "Global Deal for Nature," suggest 30% protection for each ecoregion and an additional 20% for climate stabilization. However, climate change may undermine PA effectiveness, as species assemblages and ecosystems shift due to range changes, abundance shifts, and extinctions. Terrestrial lands, including those within PAs, are experiencing increased climate change exposure and novel climatic conditions. These changes will lead to shifts in biome extents and habitat transformation. Conservation frameworks relying on static PA boundaries are challenged by this impermanence, requiring a more strategic and flexible PA prioritization approach that addresses climate change and anticipated biodiversity shifts. Prioritization schemes range from static place-based approaches (e.g., key biodiversity areas) that ignore climate change to climate space approaches focusing on climate type representation. This study examines how climate-driven changes in terrestrial ecoregions and biomes will alter representation within the global PA network and impact 30% area-based conservation targets, utilizing spatial climate analogs.

The authors review existing literature on the importance of protected areas in biodiversity conservation and the challenges posed by climate change. They discuss different approaches to conservation prioritization, ranging from static, place-based methods to dynamic, climate-space approaches. The literature highlights the limitations of static approaches in the face of climate change and the need for more flexible strategies that account for shifting species distributions and ecosystem dynamics. The authors cite several studies that project changes in biome extent and habitat transformation under climate change scenarios. They also discuss the ongoing negotiations for post-2020 global conservation targets, emphasizing the need for climate-informed prioritization schemes in the context of persistent failures to meet existing targets (e.g., Aichi Biodiversity Targets).

The study uses spatial climate analogs, a method that combines place-based and climate-space approaches, to project mid-century climate conditions under a +2°C warming scenario. It utilizes four climate variables (average minimum temperature, warmest monthly average maximum temperature, annual cumulative actual evapotranspiration, and climatic water deficit) as strong predictors of species distributions. Data sources include TerraClimate for climate variables and the World Database on Protected Areas (WDPA) for protected area information. The method identifies present-day locations (analogs) within a 2000km radius that share similar climates to those projected for a focal location in the future. Mahalanobis distance is used to quantify climatic similarity, considering interannual climate variability. Changes in the distribution and extent of biomes and terrestrial ecoregions (from Dinerstein et al. 2017) are projected. The proportion of each ecoregion is assessed for: (1) remaining climatically stable, (2) transitioning to conditions representative of other ecoregions or biomes, and (3) remaining stable and intact (low human modification). Uncertainty is addressed by considering the dissimilarity in climate between future conditions and analogs and the plurality of votes for projected ecoregion shifts. The study makes a database of climate analogs publicly available and provides a web-based tool for visualizing potential future impacts.

By mid-century, the study projects that 53.9% of global land area will experience climatic conditions corresponding to different ecoregions, and 21.8% will transition between biomes. A larger proportion (58.2%) of land area within PAs is projected to transition between ecoregions. The prevalence of transitions varies by latitude, with peaks in northern mid-latitude and southern tropical/subtropical regions, which often have low protected area coverage. Areas with low protection and high transition potential include the southeastern U.S., northern India, central Asia, central Africa, and boreal forests. High protection and high transition potential areas include the Amazon Basin and central Australia. 3.6% of total land area and 4.6% of area within PAs are projected to have regionally non-analogous climates. Biome-level transitions generally mirror those for non-protected land, with some exceptions. Tropical Broadleaf Forests, Temperate Conifer Forests, and Tundra are over-represented in the PA network, while Temperate Grasslands and Temperate Broadleaf Forests are under-represented. The average change in percent protection across all ecoregions is minimal, but larger transitions occur at the biome and ecoregion scales. Significant increases in protection are anticipated for some biomes (Tropical Dry Broadleaf Forests, Temperate Grasslands and Shrublands, and Mediterranean Forests and Scrub), while declines are anticipated for others (Tundra and Montane Grassland and Shrublands). 16.8% of ecoregions currently meet or exceed a 30% protection target, but many are projected to lose protected area. For many ecoregions, the area needed to achieve protection targets exceeds the area projected to remain stable and intact. Uncertainty analysis reveals higher climate dissimilarity in tropical and subtropical ecoregions and lower plurality of votes in regions with complex topography.

The findings highlight the substantial potential for climate change to alter the distribution of terrestrial ecoregions and biomes, significantly impacting the effectiveness of the global PA network. The projected shifts in ecoregion representation challenge existing area-based protection targets and necessitate a reassessment of conservation strategies. The study's results provide crucial information for evaluating indicators for the post-2020 Global Biodiversity Framework (GBF), particularly for assessing future losses and gains in ecosystem extent. The emergence of novel geographic patterns in biodiversity requires prioritizing schemes that enhance the resilience of the PA network. Identifying areas with minimal human modification and projected climatic stability is a starting point, but prioritizing solely on stable areas is insufficient to meet current targets. Conservation efforts should also consider investing in areas poised for change, promoting climatic connectivity between stable areas, and incorporating matrix habitats and human-modified landscapes. The study suggests using spatial climate analogs to identify stable areas as conservation anchor points and potential sources of pre-adapted biota. Combining climate analog and connectivity modeling approaches can improve strategic PA expansion.

Climate change will substantially shift the representation of ecoregions and biomes within the global protected area network. This study demonstrates the need for prioritizing schemes that consider climate-driven changes in biodiversity patterns. Focusing conservation efforts on climatically stable areas is necessary but insufficient to meet current conservation targets. A combination of protecting stable refugia and promoting connectivity between them through matrix habitats and human-modified landscapes will be essential. Future research should focus on refining and assessing the spatial climate analog approach, improving our capacity to predict and mitigate the effects of climate change on biodiversity.

The study relies on several assumptions, including that ecoregions are effective surrogates for ecosystem-level biodiversity features and that climate and biodiversity patterns are in equilibrium. Ecoregions can span broad climatic gradients, and other non-climatic factors influence biota. The study acknowledges that rates of climate change may exceed the ability of some species to keep pace, and that intensifying disturbances can further alter patterns. The study uses a specific warming scenario (+2°C) and a single ecoregion classification system. Uncertainty in climate projections and spatial analog methodology is acknowledged, and efforts are made to address it. However, the results still provide valuable insights into the potential impacts of climate change on the global PA network and inform more dynamic and resilient conservation strategies.