Soil moisture, a very spatially variable quantity, is a critical component to models ranging in application and scale from global climate change to forest carbon cycling. Such models are typically very sensitive to changes in soil moisture, due to its effect on the thermal and hydraulic characteristics of the soil and its biochemical functioning. The techniques and sampling strategy of measurements is therefore critical to effectiveness of modeling efforts, particularly with regard to spatial variability at a variety of scales.

We present analysis from a measurement program designed to examine the local to microscale spatial variability (length scales of 1 m to 1 km) of soil moisture in the Morgan-Monroe State Forest (MMSF), south-central Indiana, USA. Statistical representativeness of individual point measurements is discussed as well as physical controls of spatial variability associated with topography and other site variables. In addition, differences in measurement technique and sampling strategy are critically examined. Given the significant variability found even over small scales, where precipitation is approximately spatially uniform, the most appropriate approach to accurate and representative measurement of soil moisture in the context of instrumentation limitations is assessed.