An encouraging headline with a complicated map
A protected area can become more suitable for endangered wildlife overall while losing valuable habitat in particular locations.
A study of Northeast China Tiger and Leopard National Park offers a striking illustration.
Researchers modelled habitat suitability for Amur tigers and Amur leopards in 2017, 2021 and 2024, then combined the spatial patterns with measures of landscape connectivity.
Estimated suitable areas increased for both species, but the gains were not uniform.
The study also introduced a biodiversity-credit accounting framework to help park managers distinguish areas to protect, restore or connect.
Those credit units represent modelled ecological conditions; they are not money, verified marketable offsets or a direct count of additional animals.
Why the border region matters
Amur tigers and Amur leopards depend on large, connected landscapes and face challenges when human infrastructure fragments their habitats.
Northeast China Tiger and Leopard National Park spans a region where conservation management must account for forests, roads, settlements and cross-border ecological connections.
The research focused on the Suiyang Administration Branch, one of the park’s core administrative units and around 19.8% of its area.
Its importance lies partly in its monitoring network and its role in a landscape used by both large cats.
The research therefore addresses how to plan space for flagship species, not simply whether a park’s area is legally protected on paper.
The three snapshots of habitat suitability
Lili Han, Yaoyao Kang, Chao Wang, Zhenfeng Zhu, Youjun He and Ruirui Gao analysed information from 2017, 2021 and 2024.
The underlying species-occurrence information came from camera-trap records collected across 2017–2024, combined with environmental data.
The researchers used maximum entropy modelling, often abbreviated MaxEnt, to estimate how suitable different locations were for the two species.
They then identified areas meeting suitability thresholds, explored overlap and assessed connections among habitat patches.
This is an approach to spatial modelling, not a complete census of tigers and leopards.
A patch becoming more suitable according to a model does not automatically establish that an animal now lives there.
What changed for Amur tigers
The authors estimated that suitable habitat for Amur tigers grew from 135.27 square kilometres in 2017 to 159.52 in 2021 and 315.12 in 2024.
That is a net increase of 179.85 square kilometres across the full study period.
The pace differed between periods: the estimated net gain was 24.25 square kilometres from 2017 to 2021 and 155.60 square kilometres from 2021 to 2024.
Those values describe the area classified by the research as suitable under its model and thresholds.
They should not be translated into the number of tigers supported, an increase in the actual tiger population or a guarantee that the newly suitable patches will remain usable under future environmental and human pressures.
What changed for Amur leopards
The corresponding suitable area for Amur leopards grew from 55.52 square kilometres in 2017 to 98.76 in 2021 and 136.30 in 2024, a net increase of 80.78 square kilometres.
These estimates show an expansion in potential habitat, but they also indicate that leopards occupied a more restricted modelled suitable area than tigers within the Suiyang study region.
Habitat suitability is influenced by the environmental predictors and the occurrence records used to fit the model.
It does not independently establish reproductive success, prey abundance or stable movement through every corridor.
Conservation decisions therefore require continued observation and field interpretation alongside the maps.
The overlap between the two species
Some habitat patches were classified as suitable for both Amur tigers and Amur leopards.
The estimated overlap rose from 39.64 square kilometres in 2017 to 85.64 in 2021 and 120.70 in 2024.
The net increase was 81.06 square kilometres.
This overlap could matter because landscape planning must consider more than one species, but a shared suitability classification does not mean individual animals occupy the same patch at the same time.
The study also separated tiger-only and leopard-only suitable areas and observed that their patterns evolved differently.
That reinforces the need to avoid relying on one pooled conservation score to summarise all species-level conditions.
A good-looking map is still a model
The authors reported strong discrimination statistics for their MaxEnt models, with mean AUC above 0.9 and mean true skill statistic above 0.8 for both species in the three years evaluated.
These values indicate that the fitted models distinguished sampled conditions well under the evaluation approach.
They do not validate every predicted future movement or establish that land-cover changes were caused by park policies.
Camera traps are valuable sources of occurrence records, but their locations and detection opportunities may not cover the landscape evenly.
Environmental predictors can also change over time.
Habitat monitoring should therefore combine model evaluation with systematic field observations, especially in newly identified areas near roads and settlements.
Counting credits without treating nature like currency
The researchers developed biodiversity credit units, or BCUs, for their accounting exercise.
The framework combined the area of potentially suitable habitat, its modelled suitability and an adjustment for each patch’s role in connecting habitat.
Such an index can help managers compare changes across space and time using a consistent calculation.
It does not create an interchangeable financial instrument by itself.
A gain in BCUs is a model-defined improvement in habitat conditions, not a legal right to offset damage elsewhere and not a direct measurement of individual animal welfare.
Treating ecological quantities as accounting units can be helpful for planning, but the underlying assumptions must remain visible.
More credits in the later period
The composite biodiversity credit gain for the two species was reported as 4,860.19 BCUs during 2017–2021 and 15,099.21 during 2021–2024.
The latter was approximately 3.11 times the earlier period’s gain, giving a cumulative composite increase of 19,959.40 BCUs across 2017–2024.
Those numbers are meaningful only within the article’s particular framework; they do not convert directly to money or prove a threefold improvement in actual biodiversity.
The authors also reported separate species-level results, with higher modelled credit gains for tigers than leopards.
A single composite total can hide where the species experience different spatial opportunities and risks.
The gains did not cover every corner
The maps showed credit gains concentrated primarily in the northeastern part of the study area, while areas of loss persisted in the southeast.
Such regional contrasts are crucial for conservation.
If an overall total rises, managers might be tempted to declare broad success and reduce attention to vulnerable corridors.
The authors instead emphasised the value of examining where positive and negative changes occurred.
A small patch of habitat can have a large connectivity role even when it contributes relatively little to total area.
Restoring or maintaining a location that links otherwise separated suitable patches may matter more for movement than simply enlarging a less connected patch somewhere else.
Connectivity changes the meaning of area
A forest fragment is not necessarily useful to a mobile carnivore just because it looks suitable on a habitat map.
The ability to move among patches affects access to resources and the broader ecological function of the landscape.
The study incorporated patch-connector contributions into its credit calculation, seeking to recognise places that connect other suitable patches rather than judging value only by size.
By 2024, patches with high connector contributions for tigers were concentrated largely in the northern general-control zone and nearby areas, while high-contribution leopard patches were found in more separate northern and central clusters.
These patterns call for species-specific corridor planning rather than assuming one connectivity intervention will benefit both equally.
A conservation success may raise other pressures
The paper noted that changes near agricultural edges, settlements and low-intensity development deserve special caution.
A modelled increase in suitability on cropland does not imply that human-modified land has become stable habitat for an endangered predator.
It may reflect environmental conditions near forest edges or changes associated with landscape use.
Where large carnivores and people share space, conservation plans must also consider conflict risks, livestock protection, infrastructure barriers and practical constraints on settlement relocation.
The analysis identified such interfaces as areas requiring monitoring, not proof that more tiger or leopard attacks will occur.
A stronger corridor strategy needs to protect wildlife while accounting for the lives of neighbouring communities.
Four management priorities, seven functional zones
By combining habitat changes with modelled credit patterns, the researchers proposed four broad management tiers and seven secondary functional zones.
The tiers covered core protection, buffer consolidation, habitat restoration and connectivity enhancement, and basic conservation management.
This offered more differentiated advice than a simple protected-versus-unprotected map.
For example, maintaining already suitable connected habitat differs from restoring degraded edges or protecting narrow linking patches.
The zoning proposals are based on the study’s modelling assumptions and spatial overlays; their effectiveness was not measured through a controlled implementation.
Managers would still need to evaluate costs, land use and species responses before choosing individual projects.
What field monitoring must add
Remote sensing and distribution models can provide consistent coverage across wide areas, but they cannot substitute for all on-the-ground evidence.
Camera traps, prey surveys, movement records and conflict reports can help determine whether predicted high-suitability patches are occupied and whether individual animals can cross them safely.
Long-term observation can also identify when a newly suitable area becomes unusable after road expansion or changing human activity.
This paper supplies a framework for planning and prioritisation, not a final assessment of how many additional tigers or leopards survived because of specific management decisions.
That distinction is especially important when conservation findings inform public claims about endangered species recovery.
Connections to other landscape research
Research Today has examined related approaches in a study of habitat and climate risks for Finnish wildlife and research using local knowledge to locate Himalayan takin aggregations.
These reports concern other populations and methods and do not independently validate the Amur cat estimates.
Their wider significance is that conservation needs more than a single headline indicator.
Species occurrences, habitat condition, land use and the practical knowledge of those working in the landscape can all contribute different information to decisions about what needs protection.
The boundaries of this accounting approach
The research covers one administrative branch and three modelled time points, so it cannot establish that all of Northeast China Tiger and Leopard National Park experienced the same changes.
The credit calculation depends on chosen suitability thresholds, area weights and connectivity adjustments.
Its output is particularly sensitive to how habitat patches and links are represented.
More suitable habitat need not mean that animals are already using it, and a rising composite credit total may coexist with severe local losses.
The authors presented the framework as a methodological reference requiring continued validation and dynamic updating.
It should not be used as an automatic substitute for environmental impact assessments or management decisions involving local communities.
What the study contributes
The strongest takeaway is a detailed, spatially qualified account of potential habitat change.
Modelled suitable habitat in the Suiyang area increased by 179.85 square kilometres for tigers and 80.78 for leopards from 2017 to 2024, with overlapping suitable area rising by 81.06 square kilometres.
The researchers’ composite biodiversity-credit index increased more quickly in the later period, even as localised losses continued.
That mix of gains and vulnerabilities helps explain why protected-area success should be assessed through habitat quality and connectivity as well as total area.
The next step is to verify how well such measures track real animal movement and conservation outcomes over time.
Source Information
Study title: Biodiversity credit accounting in the Suiyang Administration Branch of Northeast China Tiger and Leopard National Park based on changes in flagship species habitat suitability
Authors: Lili Han, Yaoyao Kang, Chao Wang, Zhenfeng Zhu, Youjun He, Ruirui Gao
Journal: Frontiers in Environmental Science
Publication date: 10 November 2026
Study design: Spatiotemporal habitat suitability and connectivity modelling with a proposed composite biodiversity-credit accounting and management zoning framework
Sample or scope: Suiyang Administration Branch, Northeast China Tiger and Leopard National Park; camera-trap and environmental data for modelled habitat in 2017, 2021 and 2024
Key finding: Modelled suitable habitat grew by 179.85 km² for Amur tigers and 80.78 km² for Amur leopards from 2017 to 2024, while localised losses persisted








