The net GHG emissions of the China Three Gorges Reservoir: I. Pre-impoundment GHG inventories and carbon balance

One of the paramount questions related to environmental and climate change impacts from hydropower and reservoirs, is how to quantify the greenhouse gas (GHG) emissions of dam construction and reservoir creation, mostly in terms of reservoir net GHG emissions. Net emissions are described as the emissions after impoundment subtracting the emissions before the reservoir was built (pre-impoundment). The evaluation of pre-impoundment GHG emissions is essential to answer the above questions, yet there are few related case studies. Herein, we proposed a conceptual framework to evaluate the pre-impoundment GHG emissions of China's Three Gorges Reservoir (TGR). Reservoir flooded areas prior to impoundment were divided into two categories: 1) flooded land, where pre-impoundment CO2 and CH4 fluxes from different historical land uses were estimated following tier 1 methodology of IPCC national inventories, and 2) river surface, where pre-impoundment CO2 fluxes were estimated by a calibrated two dimensional modified biogeochemical model with an air-water gas transfer module. An empirical regression model between measured air-water CO2 and CH4 fluxes in unflooded river reaches was used to estimate pre-impoundment river surface CH4 flux. The pre-impoundment GHG emissions of the TGR were 5.1 x 10(5) tCO(2eq).yr(-1), with 95% confidence intervals of 4.7-6.1 x 10(5) tCO(2eq).yr(-1). Approximately 46% of the pre-impoundment GHG emissions were from flooded land, while the rest 54% were from river surfaces. Mass balance indicated that approximately 72% of the downstream riverine C export was from the upstream river basin of the Yangtze River. Pre-impoundment C emissions were only similar to 6.58% of the total riverine C export downstream. Most of the C in the system was mainly from the upstream river basin of the Yangtze River; thus, an increase in anthropogenic loads of C and nutrients in the Three Gorges Reservoir Area did not result in an apparent increase in pre-impoundment river surface GHG emissions. (C) 2020 Elsevier Ltd. All rights reserved.