Supported by:

Real time monitoring of grasslands, providing information for grazing management
by Mr Graham Donald, CSIRO Livestock Industries

Problem: Grassland degradation in inner Mongolia

• In recent years managing the Greater Inner Mongolian Steppes has been complicated by greater seasonal climatic variation; longer winters and warmer shorter growing seasons. Overstocking has occurred due to the lack of past experience with climatic change and coupled with financial pressures to increased stocking numbers.
• There are identifiable areas that are considered to be below marginally productivity. Immediate de-stocking of the Xilingol (Xilinhot) region is not a viable option as sustainable grazing maintains the balance between the dominant pastures species (Stipa grandes and Leymus chineses) and other less desirable species. It is critical to identify these regions.
• It is therefore critical to manage livestock stocking rate. It is believed that 70% of Inner Mongolian farmers are traditionally seasonal nomadic (transhumance) during the spring to summer period of pasture growth. Agricultural advisors of these regions are responsible for the management of the Grasslands.
• The Xilingol (Xilinhot) League of Inner Mongolia over the last few years has recorded rainfall below normal. Many Chinese climate experts suspect this dry trend is not necessarily transitory. North China is becoming warmer, drier and windier due to deforestation and global climate change.

The Chinese Academy of Agricultural Science (Dr. Xiaoping Xin, CAAS, in August, 2005) considered it necessary to examine their pasture growth GIS/spatial estimation models for environmental management and monitoring. These models ostensibly were to provide a mechanism for Government advisors to make management decisions to reduce the incidence of over grazing and consequently land degradation and desertification. Xilingol League is one of the four global and largest examples of the world grasslands.

Capability of Australian science: Real time monitoring of grasslands, providing information for grazing management

Discussions were held around the monitoring of grasslands and the management of grazing livestock. CSIRO models for grasslands monitoring utilize satellite remote sensed data and climate surfaces.

• My primary role was to examine the Chinese models and make recommendations on possible improvements that included some modification.
• Review the algorithm used and application of these models in a GIS environment.
• Provide examples on the type of data required to enable the grassland managers to apply appropriate stocking rates for sustainable management.
• Provide suggestions on improvements and scope for redesigning their models to determine the pasture status at strategic times to allow government officials regulate stocking numbers with available biomass.
• The possible use or adaptation of CSIRO’s weekly real-time pasture growth rate (PGR) and Feed on Offer (FOO) models was never brought forward as a consideration; however, their capability provided a benchmark for assistance.
• To enable real time monitoring successfully, weekly, 2 weekly or monthly maximum composited MODIS Normalized Difference Vegetation Index (greenness index, NDVI) for this region and spatial climatic (rainfall, temperature, evaporation and solar radiation) will be required. CSIRO’s Pastures from Space partner (Department of Land Information – Remote Sensing Centre) would be capable of providing satellite information.

Opportunity: Spatio-temporal modeling of grasslands, simple DSS delivery models; 'measure is to manage'

The CSIRO Pastures from Space (PfS) project delivers near real time information at the regional, catchment, whole farm and within paddock level. This information underpins the tactical and strategic decision making for Australian agricultural businesses in Southern Australia. PGR is a deterministic model providing pasture growth rate (kg DM/ha.day) and biomass (feed on offer, FOO) model provides (kg DM/ha), quantitative estimates of green biomass. Both models were designed and operated in a GIS environment.

• The relationship between NDVI and fAPAR (fraction of absorbed photosynthetic active radiation) provides the fundamental basis for estimating PGR remotely and this estimate is improved by including climate; both as spatial and temporal component.
• Both PGR and FOO models would require extensive research including modifying, calibrating and validating before being deployed in this different grasslands environment.  Also develop an extensive field pasture assessment protocol.
• Provide real-time weekly assessments of PGR and FOO to farmer and Government advisors in-conjunction with appropriate software (Pasture Watch – Fairport Technology) as a means to improve long and short term sustainable productivity by managing pasture utilization with some degree of confidence.
• Provision of historical PGR’s and some further understanding of the affect of recent climate extremes would be of assistance.
• Provide FOO (Biomass) maps of critical areas to define areas of risk.
• Develop a commercial system to integrate grasslands management and livestock productivity.

http://www.pasturesfromspace.csiro.au
http://www.fairport.com.au