Work on a variety of spatial scales is needed to analyze structural and functional proper ties of ecosystems and detect change. The methodologies involve field studies at the meter scale or less, use of models, remote sensing, and spatial and temporal analyses. Research begins at the single plant level and continues up to the patch scale to understand biogeochemical and physiological change. At the watershed to landscape level, influences on soil properties, biogeochemical cycles, distribution of plants, and trace gas fluxes be come relevant. Patch scale and regional scale changes in ecosystem properties can then be incorporated into couplings with atmospheric models, and, at the global scale, issues such as net primary productivity (NPP) are relevant to a variety of models.
A series of databases is developed into a modeling system for vegetation, soils, elevation, etc., and incorporated in nested way coupled through a series of models (see Figure 12.1). This information can then be useful in dealing with proximate causes of change that influence gain or loss of ecosystem or biome types. For example, it can provide information for reforestation or conservation research programs, such as those that return agricultural lands to grasslands.
Natural succession and change need to be incorporated in assessment frameworks as well. Coupling of ecosystem processes related to change includes deposition and inputs of nitrogen (N), changes in acidity (pH) of rainfall, other inputs to the system, and biogeochemical and hydrological cycles of the ecosystem. Changing land cover modifies a number of ecosystem properties.
Land Use Change Impacts:
An example of some of these changes in ecosystem properties occurred in Texas with the expansion of mesquite at the time of European settlement. A succession from subtropical grassland areas to open savanna resulted from alterations in the grazing pattern and subsequent fire regime modification. This led to a number of changes in the ecosystem involving allocation of carbon, structure, albedo, energy balance, and fertility (mesquite is a nitrogen fixer so islands of fertility developed). Nitrogen trace gas fluxes were greatly altered; there were low N fluxes compared to tree and woodland areas.
Structural change can also have a great influence on hydrological cycles. Without forest cover, rainfall over the annual cycle is reduced by approximately 60%. Other hydrological effects include albedo changes, evapotranspiration changes, increased runoff, erosion, and changes in the amount of water available in the region.
An illustration of dramatic change brought about by land cover change occurs on the Oklahoma panhandle (Roger Pielke, personal communication). Under current land use practices, native rangeland vegetation has been largely replaced with irrigated land, urban area, and dryland agriculture. Even in this semi-arid system, humans have impacted land cover quite significantly. Figure 12.2a shows a native vegetation land surface parameterization for a 70 km x 100 km region. The simulated results show the pattern of clouds and convective storm activity under the native vegetation regime. Figure 12.2b repeats the simulation but with current land cover conditions for the same date. The dramatic build up in mesoscale climate activity in the region is apparent. Observed meteorological results were very similar to modeled change.
It is hypothesized that the change in land cover is responsible for creating the observed significant variation in natural weather patterns with dramatic implications. If this hypothesis is correct, then changes in landscape patterns over the past 100 years have caused meso-meteorological effects, and may have contributed to altered climatology of the region. In addition to these rapid changes, there are biogeochemical changes in allocation patterns of nutrients, and other less obvious changes.
Pristine grassland areas worldwide, especially in Asia, are changing from species-rich ecosystems with the ability to respond to changes to monocultural systems characterized by increases in soil respiration, erosional losses, and changes in secondary productivity related to domestic animals. Intensity and time of grazing impact semi-arid areas in a number of ways. Changes in fire regime cause major changes in the structure of an ecosystem, as well as its biogeochemical and hydrological cycle.
The CENTURY model can be used to study the impacts of climate change on ecosystem processes. Figure 12.3 demonstrates how the model integrates a water budget model, a grass/crop model, a soil organic model, and a forest model. Management practices such as irrigation, harvesting variables, input of plant residues, tillage, nutrient enhancement, modification of fertilizer rates, etc., are incorporated into the CENTURY model as well, and manipulations can reflect common land use changes.
The CENTURY model can be regionalized by parameterizing, validating for the site-specific parameterization, and incorporating regional information including remote sensing data. Remote sensing data is used as input but also to look at validation of model output for the region by looking at changes in NDVI or greenness for the growing season. This is a critical factor in assessment. It is also necessary to track results through several years to capture interannual variability.
It is possible to predict net primary productivity (NPP) and biomass against NDVI signa tures using AVHRR data. Using regional aggregation satellite data to provide NDVI and biomass provides a way to look at large grassland areas, and bring results up to the regional scale.
Microbial processes can be studied at the soil level and aggregated up to the patch and global levels. At the soil surface there are microbial changes such as growth of blue-green f-fixing algae after a burn. These are a nitrogen-fixing set of organisms that replace N into the system. After burning, there is an enhancement of N inputs through this process.
Chamber techniques are being used to look at fluxes of trace gases over a 15-year observation period. Tower techniques are being used to study CO2 and energy at the FIFE site (Manhattan, Kansas) and ammonia fluxes in burned areas of tall grass prairie. Smaller towers are being used to study methane fluxes in the boreal forest.
There has been a dramatic change in methane uptake between areas of native grassland and current wheat crop systems. The resulting reduction in the sink strength can contribute to the increase in methane in the atmosphere.