Modelling soil greenhouse gas exchange of the Haean catchment
TERRECO WP 2-15
Von 03/2009 bis 07/2015Projektleiter
: Ralf Kiese
, Klaus Butterbach-Bahl
, John Tenhunen
: Young-Sun Kim
: IRTG 1565 WP II
TERRECO - Complex Terrain and Ecological Heterogeneity - Evaluating ecosystem services in production versus water yield and water quality in mountainous landscapes
Abstract 2011: Agriculture practices such as crop rotation, tillage and fertilization have significant impact on ecosystem nutrient turnover and losses with potential harm to the atmosphere (greenhouse gas emissions - GHG) and hydrosphere (leaching). The Landscape-DNDC model, originated and further developed from the DNDC model, is capable of predicting soil carbon and nitrogen turnover and associated losses of greenhouse gases and nutrient leaching from terrestrial ecosystems. The study region is located in the Haean Catchment, northeast of Yanggu County, Korea. Highland agriculture is the dominant agricultural practice in this region. The Landscape-DNDC model was applied to estimate nitrogen turnover rates and N2O emissions as well as NO3- leaching rates of forest and various agricultural sites experiencing different management practices of Haean Catchment. Field measurements of TERRECO such as N2O emissions, inorganic soil nitrogen stocks (NH4+-N and NO3--N) as well as NO3 concentration in soil water from upland fields and rice paddies, were used for comparison with the model simulations. In particular, simulation results were compared to the measured N2O emissions from a radish field with different fertilization regimes. The underestimated simulation results indicate the need for improvements in the model, considering various agricultural practices such as heavy fertilizer use, soil dressing, and mulching and application of plastic ground covers, which were previously never considered in the model. In addition, detailed information for nitrogen fertilizer additions is required to achieve better simulation results. Simulated annual losses of nitrate in seepage (40-50 kg N) were one order of magnitude higher than N2O emissions (1.5-3.5 kg N), which indicates that eutrophication is probably an even greater concern than GHG emissions from the agricultural system in the Haean Catchment.
Keywords: Landscape-DNDC model, agriculture practices, agricultural soils, N2O emission, inorganic nitrogen stocks, nitrate leaching
Abstract 2013: Process-based biogeochemical models can be used to predict the impact of various agricultural management practices on plant nitrogen use efficiency and nitrogen losses to the environment, such as greenhouse gas emissions and nitrate leaching, by analyzing the interactions between management practices, primary drivers such as climate, soil properties, crop types and biogeochemical reactions. In this study we applied the Landscape-DNDC model, which combines functions from Agricultural-DNDC and Forest-DNDC, for simulation of C and N turnover, GHG emissions, nitrate leaching, and plant growth for Korean arable fields cultivated with radish (Raphanus sativus L.) and soybean (Glycine max) in Haean Catchment.
The annual average temperature in Haean is ca. 8.5°C and the annual precipitation is ca. 1,500 mm. According to farmers practice, the radish field received a basal fertilizer application of ca. 190 kg N ha-1 before setting up four fertilizer treatments, i.e., additional 50, 150, 250 and 350 kg N ha-1. All N treatment plots were tilled a week after application of specific N fertilizer in order to make rows and interrows. Just before radish seeding, rows were covered with black plastic mulch which was removed after harvest. In spite of the widespread usage of black mulch in Korea or even Asia, so far biogeochemical models do not consider impacts of mulch on soil environmental conditions and soil biogeochemistry. Thus, based on field measurements, we adjusted our input data and used only half of the annual precipitation and 90% of the maximum temperature as drivers for simulation of soil environmental conditions of rows with foil coverage, whereas the actual weather data was used for interrow simulations. Simulated soil moisture and temperature as well as N2O emissions and soil nitrate concentrations agreed well with field measurements. However, peak emissions after fertilization were slightly underestimated in row and interrow. Annual N2O emissions of the fertilizer treatments increased with increasing fertilization rates from around 2.1 to 2.4 kg N ha-1 in rows and 2.4 to 3.2 kg N ha-1 in interrows. Due to the sandy soil conditions and the monsoon character of rainfall, major N loss of the agricultural systems was in form of nitrate leaching. Nitrate leaching rates were as high as 290 - 450 kg N ha-1 for the different fertilizer treatments and rows and interrows. Only slight increase of measured and simulated yields (4.4 to 5.6 t DW ha-1) with higher rates of N fertilization indicate that actual farmer practices can be improved. Optimizing agricultural management considering the specific climatic and soil conditions has a high potential to reduce environmental impacts of crop cultivations in the study area.
Model simulations for a soybean field without fertilization and one tillage before seeding were also in good agreement with field observations. However, simulated N2O emissions (0.85 kg N ha-1) and nitrate leaching (54.7 kg N ha-1) were much lower than for the radish field.
Key words: Landscape-DNDC, N fertilizer, plastic mulch, N2O emissions, nitrate leaching, biomass
Poster January 2013