There is a very important research contribution that is relevant to the effect of the siting of surface temperature instrumentation with respect to the assessment of spatially representative long-term trends. The implications of her study goes beyond that of cattle grazing, as this is the first study, to my knowledge, of the effect on surface temperatures of vegetation/soil patterns on the scale of a few meters. This, of course, assures that the overlying atmosphere is essentially identical such that any differences in the soil temperatures, and other climate metrics are due to the details of the immediate land surface characteristics.
The study is in an M.S. thesis (which is a peer reviewed contribution) under the direction of Peter D. Blanken of the University of Colorado in Boulder.
The thesis is
Wolchansky, Jennifer E. (M.A., Geography)
Effects of Simulated Grazing on Soil Temperature, Moisture, and Respiration on a Shortgrass Steppe in Northeastern Colorado. University of Colorado, 2005.
The abstract reads
“Cattle grazing, a common form of land use on grasslands, could alter the climate at the surface. This study focused on how physical landscape changes associated with grazing could have a significant impact on soil temperature and moisture, and subsequently on soil respiration on a shortgrass steppe in northeastern Colorado.
Using micrometeorological data, the objective of this research was to discern the connections between energy, water, carbon and land use, particularly grazing.
To achieve this objective, the research was divided into two sections, each with its own objectives and results, yet both related to the overall objective. The first section examined the impacts of simulated grazing on soil temperature and moisture. Plots with varying fractions of bare ground were used to simulate grazing density. Results indicated that both soil temperature and soil moisture increased with an increase in the fraction of bare ground. In addition, this section investigated the effects of the arrangement of bare ground on soil temperature and moisture. Soil temperature was found to be similar under contiguous and mosaic patches of bare ground while soil moisture was found to differ.
The second section of the research expanded upon the results found in the first section. Having found the relationship between the fraction of bare ground, and soil temperature and moisture, simulated cattle grazing was indirectly linked to changes in soil respiration. Soil respiration increased with an increase in soil temperature, until a critical threshold temperature was reached. An increase in soil moisture resulted in a decrease in soil respiration.
Because grasslands cover nearly half of the earth’s terrestrial surface, it is important to understand how land management on these grasslands can affect the ecosystem. In addition, determining how grazing can influence the release of CO2 to the atmosphere is crucial at such a time when there is growing concern about global climate change.”
One of the predictions which she confirmed in her study was that
“Configuration of patchiness will impact spatial variation of soil temperature and moisture. An increase in soil temperature and a decrease in upper profile soil moisture is expected with distance from vegetation. Therefore, a plot of half bare ground and half vegetation will have a higher maximum temperature and lower minimum soil moisture than that of a randomly arranged plot.”
An excerpts in the conclusion, that is directly relevant to the issue of spatial representativeness of long-term temperature trends read
“The first phase of this research found that while Ts under bare ground was generally greater than that under vegetation, more importantly, Ts values increased as the fraction of bare ground increased under both bare and vegetated cover throughout the full period. Essentially, because there is a positive correlation between grazing density and the fraction of bare ground (Teague et al. 2004, Metzger et al. 2005), these results imply that an increase in grazing density will increase Ts. Additionally, greatest differences in Ts between bare ground and vegetated cover were found in the plot with roughly half vegetation, half bare ground. Although recommended grazing densities generally result in 15-20% bare ground (Justin Derner, personal communication), this result indicates that if an area were overgrazed enough to produce 50% bare ground, then implications would be great. Such a significant difference in Ts between the bare and vegetated cover could influence surface heat fluxes and, in turn, could affect atmospheric conditions.”
In terms of station siting, the spatial assessment completed by Jennifer Wolchansky and Peter Blanken can be interpreted as what occurs if a single location undergoes these local land surface changes over time. The effect on the surface temperature trends as more bare ground is gradually exposed for whatever reason would result in a warming trend that can be misinterpreted as a larger scale surface temperature trend. Similarly, a gradual reduction in bare soil coverage due to revegetation, would result in a cooling trend, which could similarly be misinterpreted as a larger scale cooling.