Ecological Indicators (Mar 2024)
Trade-offs in the use of direct and indirect indicators of ecosystem degradation for risk assessment
Abstract
Ecosystem risk assessments estimate the likelihood of major transformations (ecosystem collapse) over a specified time frame. They require an understanding of the biotic and abiotic processes that drive declines. Relative Severity and Extent of Decline quantify essential dimensions of ecosystem degradation as part of the International Union for the Conservation of Nature (IUCN) Red List of Ecosystems risk assessment protocol. These flexible and powerful concepts are operationalised through ecosystem-specific indicators of functional decline. Here, we examine trade-offs in risk assessment between direct, yet data-demanding indicators and indirect indicators that are more widely applicable with global data sets. Using a case study of multiple tropical glacier ecosystems, we compared estimates of risk based on a direct indicator of functional decline (ice mass) with those based on an indirect indicator (bioclimatic suitability). The direct estimate of Relative Severity was based on the projected changes in ice mass using a glacier ice mass balance and dynamics model, while the indirect estimate was calculated from the expected changes in suitability based on a correlative habitat suitability model parameterised with current occurrence records. For reference, we calculated probability of ecosystem collapse from simulations of the ice mass balance and dynamics model. We found that the indirect indicator systematically underestimated risks of ecosystem collapse compared to the direct indicator and returned a different rank order of risks across glaciers due to prominent discrepancies in some units. Small and isolated glaciers located outside the tropical Andes are uniformly exposed to high levels of degradation and have high probabilities of collapse before 2080, whereas tropical Andean glaciers exhibit different rates of degradation, but are expected to undergo very severe degradation before 2100. For these larger units a detailed analysis of spatial differences in future projections could inform regional and local strategies for future monitoring, management and conservation action that can benefit people and nature. Evaluating Relative Severity and Extent of Decline over time and with different ecosystem-specific indicators allowed us to describe trends across a group of functionally similar ecosystem types and compare their performance in assessment units of different size and risk of collapse. The methods could be applied to other ice or snow-dependent ecosystems, while the case study should be instructive for development of risk indicators in many other ecosystem types.
