Ecological Indicators (Sep 2021)
A prioritization metric and modelling framework for fragmented saltmarsh patches restoration
Abstract
Saltmarsh is a coastal ecosystem providing crucial ecosystem services, and its continued degradation and fragmentation has drawn increasing attention. However, how to effectively restore the connectivity between fragmented saltmarsh patches remains an open challenge. In this study, we developed a metric and modelling framework that prioritised saltmarsh patches for restoration. To demonstrate our approach, we simulated spatially explicit restoration schedules for Suaeda salsa patches at the Yellow River Delta National Nature Reserve, China, using three strategies: increasing-patch-area, increasing-number-of-patches and a benchmark unrestrictive prioritization strategy. We prioritised patches for restoration based on a number of widely used graph-theoretic landscape connectivity and metapopulation capacity metrics. Our simulation results suggested the rank connectivity-importance of extant patches was correlated within the group of graph-theoretic connectivity metrics or metapopulation capacity metrics, but unrelated across group. The unrestrictive prioritization strategy clearly outperformed the strategies of increasing-patch-area and increasing-number-of-patches which returned comparable connectivity restoration outcomes. For the more effective unrestrictive prioritization strategy, there were substantial differences in the simulated priority patches between metrics that considered stepping stone effects and those did not. While the former resulted in corridor-building priority patches that led to a more connected landscape throughout the region, the latter led to local clustering. We recommend use of the total probability of connectivity (PC) among the metrics we tested due to similarity of results to other metrics and its simulation efficiency. The proposed framework is readily applicable to prioritise areas for connectivity conservation and restoration in any monospecific ecosystem at the regional scale.