Reducing the likelihood of carbon loss from wetlands by improving the spatial connections between high carbon patches

Abstract With stores that are more than 50 times greater than in terrestrial ecosystems, riverine and coastal ecosystems are the most important carbon (C) stocks worldwide, with most C stored in wetlands. Wetlands are, however, among the most threatened ecosystems worldwide, and face pressures from hydropower development, flow regime alteration, and land use/cover change (LUCC). Despite great efforts to protect and restore wetlands, anthropogenic drivers (e.g., land reclamation) still result in more C loss than gain in most cases. Of these, LUCC is driven by social and economic processes, and is a self-organizing, path-dependent phenomenon. Therefore, we proposed a land use management framework with which to analyze the spatial patterns in the drivers of LUCC, and then verified the strategy using two study areas in the lower reaches and delta of the Yellow River. The first step of the process was to analyze LUCC from historic land use/cover maps (from 1995 to 2015 for the lower river reaches and from 1970 to 2015 for the delta), and identified a suitable land use (waterbody) for the connections. We calibrated and validated the DINAMICA model using data for changes in C. We then set up different scenarios and created connections between high C patches. The simulation results showed that even slight modifications in the connections with water could trigger noticeable changes in the spatial patterns of C gain and loss, and that original hotspots of C loss could be converted to areas of C gain in some cases. Our findings highlight the need to consider both spatial patterns and drivers of LUCC when protecting wetlands and show that water-sediment regulation in the Yellow River should be coordinated with dynamic changes in the landscape in the lower reaches and delta.