Modeling riparian flood plain wetland water richness in pursuance of damming and linking it with a methane emission rate

Abstract

Modeling water richness in dam-induced riparian wetland and linking methane (CH4) emission rate has immense importance to understand the regime of methanogenesis. Multi-model machine learning approaches have been adopted for developing water richness (WR) models and an equation-based method has been followed for estimating CH4 emission from the wetland. REPTree model is found to be the best representative WR model. About 194.14 km2 wetland areas have been squeezed; very high water richness area has been reduced from 47.37 to 29.59% in the post-dam period. Along with CH4 emission has also been reduced by 52.61%. The high CH4 emission rate is attributed to the higher WR area. Linking CH4 emission with WR demonstrates a positive relation (0.59 for pre-dam and 0.23 for post-dam at >0.001 level of statistical significance) establishing the very vivid effect of WR on CH4 emission.