Controls of carbon flux in a semi-arid grassland ecosystem experiencing wetland loss: Vegetation patterns and environmental variables

Abstract

Inner Mongolia, China, contains the world’s largest grassland ecosystem. This has many areas of wetland, which provide important ecological services, especially carbon sequestration in the semi-arid terrestrial ecosystems. However, the area of wetland has decreased sharply in the past two decades. This study examined ways to recognize and extract wetland from grassland to determine the difference among net ecosystem exchange (NEE), gross primary production (GPP) and ecosystem respiration (Reco) between wetland and grassland and to evaluate the influence of wetland loss on carbon sequestration in the grassland. The eddy covariance (EC) flux technique was coupled with the Vegetation Photosynthesis and Respiration Model (VPRM) to upscale the spatial patterns of carbon flux. The results showed that the region was a carbon source in 2016, probably caused by overharvesting and degradation of forests. The value of the NEE (average: 6.63 ± 3.50 g C m−2 d−1) in grassland was apparently higher than that in wetland (average: 0.86 ± 1.69 g C m−2 d−1), which suggested that the capability of carbon sequestration in wetland was still stronger than that in grassland even in carbon loss condition. The results showed a positive relationship between aboveground biomass (AGB) and ground-based daily GPP or Reco for both wetland and grassland and a negative relationship between AGB and NEE. The ground-based daily NEE was also significantly related to soil water content (SWC) but showed no relationship with daily precipitation (PRE), which suggested that SWC was a more important impact factor than precipitation on CO2 flux exchange in the study area. The change between wetland and grassland did not influence the positive relationship between AGB or SWC and CO2 flux. Our study provides a new way to determine the spatial CO2 flux exchange and its controlling factors (environmental variables and vegetation patterns) and to successfully analyze its differences in wetland and grassland.