Modeling SOC and NPP responses of meadow steppe to different grazing intensities in Northeast China

Abstract

Grassland ecosystems play an important role in Chinese terrestrial ecosystems. However, great demand and excessive utilization of human beings on grassland resources have made it more susceptible to rapid degeneration in ecosystem properties and soil carbon levels. Among them, grazing is one of the key factors to make grassland ecosystems degraded. Therefore, it is very important to graze sustainably on grassland for preventing the degradation of grasslands. In this paper, as a case study, we simulated the potential changes of soil organic carbon (SOC) and net primary productivity (NPP) of meadow steppe dominated by Leymus chinensis under different grazing intensities based on CENTURY model (V4.0) in order to evaluate the effects of different grazing intensities and to pursue for optimal grazing intensity. The results showed that NPP and SOC of meadow steppe dominated by Leymus chinensis were very sensitive to grazing intensity. The optimal grazing intensity could be expressed by live shoots, and it should be less than 40% of whole live shoots per month, in order to mitigate the degradation of Leymus chinensis grassland and maintain its sustainable development.

Introduction

Grasslands are one of the most widespread vegetation types, covering approximately 40% of the land surface (Frank et al., 2000, Glenn et al., 1993, World Resources, 1986) and containing approximately 30% of global soil carbon stocks (Anderson, 1991). Ranging from the savannas of Africa to the North American prairies and the converted grasslands of Latin America and South-East Asia, these ecosystems play a very important role in regulating global carbon cycle by plant and soil organic matter (SOM) in grasslands and providing meat and grain for human beings.

However, with the development of industrialization and the increase in population, great demand and excessive utilization of human beings on natural resources have made grassland ecosystems more susceptible to rapid degeneration in ecosystem properties and soil carbon levels (Archer et al., 1994, Ojima et al., 1994). The decrease, degradation and exhaustion of grassland resources have been universal. Among them, grazing is one of the key factors to make grassland ecosystems degraded. Carbon, which is translocated from plant parts to herbivores, is usually 25–50% or more of aboveground net primary productivity (NPP) (McNaughton, 1976, McNaughton, 1985, Lauenroth and Milchunas, 1992) and perhaps 25% of the belowground productivity (Coleman, 1976, Ingham and Detling, 1984, Lauenroth and Milchunas, 1992). These changes would not only contribute to the increase of “greenhouse gas”, especially CO₂ concentration in the atmosphere, but also affect surface temperatures through boundary changes in vegetation cover and alter significantly the biogeochemistry, especially carbon cycling, through affecting NPP and soil carbon storage of grassland, and consequently may lead to feedback to components of climate change (Burke et al., 1991).

China is one of the countries in the world with the fast industrialization and urbanization during recent two decades. Due to large-scale reclamation of grassland and overgrazing as well as climate change characterized by ‘global warming’, the area of degraded grassland in China amounts to 8.667 × 10⁷ h m², and occupies about one third of the available grassland area. The desertification area at present reaches about 2.622 × 10⁸ h m², occupying about 27.3% of the land (Tian and Ma, 2001).

Leymus chinensis grassland distributed in the east part of Euroasian continent grassland region plays a very important role in temperate grassland ecosystems. It is a kind of zonal grassland type, and occupies about 4.2 × 10⁷ h m² in the world. Among them, more than half is located in China (Wu et al., 1995). As a typical grassland type in China, it is a very important base of livestock farming in northern China. With the increase of population and livestock, the degradation of Leymus chinensis grasslands has been serious. The area of degraded grassland at present makes up about 50% of the total available grassland area in Inner Mongolia, and nearly 20% of it is seriously degraded grassland. In the famous Hulunbeir and Xilingole steppes, the areas of degraded and serious degraded rangeland reached about 23 and 41%, respectively (Li et al., 2002). The conflict between grazing demand and grassland enduring capacity has been a serious challenge to human being.

Studies of the effects of grazing on above ground productivity and carbon sequestration have generated varying results (Milchunas and Lauenroth, 1993). Grazing by domestic herbivores has been shown to reduce both primary productivity and C sequestration for a mid- and tall-grass community in the North American Great Plains (Derner et al., 2006) and for Festuca swards in southwestern Alberta (Dormaar and Willms, 1998). However, grazing does not always reduce soil organic carbon and the effects may vary with stocking rate (Schuman et al., 1999, Schuman et al., 2001) and nitrogen application in ecosystems (Mortenson et al., 2004). In European grassland sites, the net carbon storage also tends to increase when the amount of carbon removed by cutting and grazing is reduced (e.g. extensive management) (Milchunas and Lauenroth, 1993). Therefore, it is very important to estimate an appropriate grazing strategy in order to maintain a sustainable grassland ecosystem.

During the past two decades, a number of simulation models about grassland grazing systems had been developed, which encompass a variety of complexity, processes, data needs and data availability (Wright and Dent, 1969, Innis, 1978, Parton et al., 1987, Hunt et al., 1991, Coughenour, 1992, Thornley, 1998, Rabi et al., 2000, Britta Tietjen and Florian Jeltsch, 2007, Badini et al., 2007). However, relatively few of these models have been rigorously tested across a range of environmental conditions. The need and desirability for model validation and sensitivity analysis has been repeatedly stressed in the literature, but most models have been tested over a limited set of conditions (Gilmanov et al., 1997).

CENTURY model, as a ecosystem-level biogeochemical model of plant-soil nutrient cycling, had been successfully applied and tested across seasonal and long-term dynamics of plant production, decomposition and nutrient cycling in various grassland ecosystems in North and Central America, Africa, Europe, and Asia (Parton et al., 1993, Parton et al., 1995, Xiao et al., 1995, Gilmanov et al., 1997, Ardö and Olsson, 2003, Wang et al., 2007). And Overall performance of CENTURY in predicting SOM dynamics is comparable with other soil organic models such as RothC, CANDY, DNDC, DAISY and NCSOIL (Smith et al., 1997). Therefore, as a case study, we would simulate the potential changes of soil organic carbon and NPP of meadow steppe dominated by Leymus chinensis under different grazing intensities based on CENTURY model (V4.0). The objectives of this study were (1) to analyze the responses of SOC and NPP to different grazing intensities. (2) To establish an appropriate grazing intensity for Leymus chinensis meadow steppe in Songnen plain.