The Redistribution of Soil Organic Carbon and Nitrogen and Greenhouse Gas Production Rates During Reservoir Drawdown and Reflooding Maren Oelbermann 1 and Sherry L. Schiff 2 Abstract: Approximately 50% of the carbon (C) stored in forest ecosystems, on a global scale, is located in the boreal forest. Inundating boreal forest soil, for hydroelectric power production, changes soil organic matter decomposition and C transformation. Soil from three reservoirs, differing in their vegetative composition and soil organic C (SOC) stock, was collected at the Experimental Lakes Area, Canada, and incubated under drawdown and reflooded conditions. Soil organic C and soil total N concentrations (g kg j1 ) and soil C 13 C and C 15 N(°) were significantly different (P G 0.05) between soil horizons within each reservoir. After 5 years of episodic flooding, there was a significantly greater SOC and N stock in the reservoirs compared with undisturbed soil. Flooding also resulted in a redistribution of SOC and N within the soil profile. CO 2 and CH 4 production rates were significantly greater when the soil was reflooded, and the highest CO 2 and CH 4 production rates came from the LFH horizons compared with a charred layer and the mineral soil. Although flooding led to the redistribution and a greater accumulation of SOC and N in the charred and mineral soil layers com- pared with undisturbed soil, the CO 2 and CH 4 production rates were lower from this part of the soil profile compared with the LFH layers. This suggested that the redistributed organic material was of lower quality compared with that of the LFH horizons. Key words: CH 4 , CO 2 , decomposition, organic matter, soil organic carbon, soil total nitrogen. (Soil Sci 2010;175: 72Y80) N early 50% of the carbon (C) in forest ecosystems, on a global scale, is located in the boreal forest, and the greatest proportion of this C is stored in the soil and detrital pool (Preston et al., 2006). Differences in climate and other biophysical char- acteristics such as the composition of vegetation, forest stand age, topography, soil type, and disturbance history lead to var- iations in total C stocks and/or its distribution within this forest ecosystem. Substantial changes in the boreal forest’s physical environ- ment, as a result of natural and anthropogenic disturbances, likely influence processes of organic matter (SOM) decompo- sition and C transformation (Tate et al., 2006). For example, anthropogenic activities, such as inundating forest soil for hy- droelectric energy production, can influence the forest’s bio- physical environment considerably (Jugnia et al., 2006). Apart from their recognized disruption in water flow and compromised water quality, the creation of large-scale reservoirs also disrupts soil biogeochemical cycles as a result of nutrient release from vegetation dieback and an increase in soil microbial activity and soil microbial characteristics (Unger et al., 2009). This leads to an increase in the cycling of C and, as such, greenhouse gas (GHG) emissions, which has been the subject of intense re- search during the past decade (The ´rien and Morrison, 2005). For example, in recent studies, Jugnia et al. (2006) and Guerin et al. (2008) found that flooded boreal forest soil is a signif- icant source of CO 2 and CH 4 . However, N 2 O emissions from boreal forest reservoirs and their contribution to global warming are minor (Hendzel et al., 2005). In 2001, the Intergovernmental Panel on Climate Change addressed the main factors affecting GHG emissions from reservoirs, including reservoir residence time and age, type of soil inundated, quantity and quality of flooded biomass, and climate and operational conditions including the length of the drawdown (DD) phase (Watson et al., 2000). Drawdown in hydroelectric reservoirs helps to increase the efficiency of power production during the winter months in northern climates. Annually, reservoir DD results in several thousands of hect- ares of periodically unflooded land in northern latitudes (Graham-Rowe, 2005). The ´rien and Morrison (2005) sug- gested that during the first decade of flooding, the major C source for CO 2 and CH 4 may be derived from the decom- position of flooded vegetation and C stored in SOM. However, seasonal changes in water depth as a result of reservoir DD may lead to a continuous supply of organic matter that under- goes decomposition in addition to changes that may occur in the soil profile because of the lateral transport of soil organic C (SOC), the input of organic matter from vegetation dieback, and inherent spatial variability of SOC (Tremblay et al., 2005). During phases of DD, new vegetation also colonizes the banks of the reservoirs, further contributing to potential sources of decaying organic matter (Graham-Rowe, 2005). Tremblay et al. (2005) also suggested that reservoir DD from autumn to spring may lead to greater and/or continual GHG production rates once they are reflooded (RF) during the summer. To date, most studies evaluating GHG production rates in northern hydroelectric reservoirs have evaluated CO 2 and CH 4 emissions during the flooded phase (Duchemin et al., 1995; Kelly et al., 1997; St. Louis et al., 2000; Soumis et al., 2004; The ´rien and Morrison, 2005; Tremblay et al., 2005; Jugnia et al., 2006; Roehm and Tremblay, 2006). This topic is currently the focus of increasing attention in the research community as well as the private energy sector because reservoirs constructed for the production of hydroelectric power have been perceived as a carbon-free source of energy. Other studies have focused on iden- tifying sources of mercury to reservoir food webs (Montgomery et al., 2000), and only a few studies focused their attention at balancing GHG emissions with changes in soil C stocks (Houel et al., 2006) and on quantifying the accumulation of C in reser- voir sediments (UNESCO, 2008). However, we are not aware of any studies to date that have evaluated CO 2 and CH 4 production TECHNICAL ARTICLE 72 www.soilsci.com Soil Science & Volume 175, Number 2, February 2010 1 Department of Environment and Resource Studies, University of Waterloo, Waterloo, Ontario, Canada. Dr. Maren Oelbermann is corresponding author. E-mail: moelberm@uwaterloo.ca 2 Department of Earth and Environmental Sciences, University of Waterloo, Waterloo, Ontario, Canada. Received July 23, 2009. Accepted for publication December 4, 2009. Copyright * 2010 by Lippincott Williams & Wilkins, Inc. ISSN: 0038-075X DOI: 10.1097/SS.0b013e3181ce0453 Copyright @ 20 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 10