Dissolved organic carbon in tropical watersheds: Linking field observation and eco-hydrological modelling

Abstract

Dissolved organic carbon (DOC) is a general description of the organic material dissolved in water. DOC is an important source of energy, carbon, and nutrient transfers from terrestrial to aquatic ecosystems. The export of DOC into aquatic ecosystems may contribute to the carbon balance of terrestrial ecosystems and to water degradation. Ongoing climate and land cover changes will affect both DOC generation and transport, with implications for both terrestrial and aquatic ecosystems. An assessment of land use land cover and climate variability’s impacts on DOC export is needed for better management of ecosystems. Watersheds are fundamental units of ecosystem functioning and are therefore an interesting organizational unit when used to understand the combined effects of land use land cover and climate variability on DOC export. Some studies have been conducted to explore this impact of land cover and climate variability on DOC, but most were conducted in a temperate environment and few in a tropical environment. In this regard, this dissertation focused on the impact of land use land cover and climate variability on DOC mobilization and export in the Rukarara River Watershed (RRW), Rwanda. The main aim is to determine how different carbon input and output processes interact under climate and land cover variability to impact DOC emanating from tropical watersheds. Data used for this study include land cover maps produced from satellite imagery, daily air temperature and precipitation, digital elevation models (DEMs), water stage, flow, net primary productivity (NPP), soil properties such as total organic carbon (TOC), total nitrogen (TN), cation exchange capacity (CEC), aluminum (Al), iron (Fe), and soil texture within the RRW. Field observations were used to quantify riverine DOC loads, soil water extractable organic carbon (WEOC), DOC in percolation water (pDOC) and leached DOC (LDOC) and to describe their spatial variation and relationships with the aforementioned factors. Statistical models (including simple and quadratic regressions, general linear model, linear mixed effect models) were used to predict DOC within the study area. An eco-hydrological model, the Regional Hydro-Ecological Simulation System (RHESSys), was used to simulate streamflow and link it with stream DOC within the study area. The results of this study show that land use land cover and climate change interact to produce soil WEOC, from which a significant fraction is transported into streams, mainly through overland flow and loaded by the Rukarara River. The riverine DOC loss was low compared to the NPP of the RRW, but may affect the function of both land and water resources with the study area. The RHESSys model detected the response of the watershed to climate variability within the RRW and captured the significant monthly variability in streamflow within the RRW. This result indicates the potential use of RHESSys to estimate streamflow in the RRW and similar tropical watersheds. Stream DOC concentration was explained by simulated streamflow in the natural forest, indicating the potential use of RHESSys model simulated streamflow to predict stream DOC in the study watershed and similar ecosystems. Further studies should evaluate the performance of the RHESSys model to simulate other hydroecological processes in the tropical environment.