Evaluating Potential Indicators of Ecosystem Processes Across Local Gradients in a Temperate Grassland.

Science-based information is needed to identify indicators of ecosystem services that may then be used to monitor natural resources and quantify effects of management.  Here our aim was to perform a local gradient study to elucidate correlative associations between vegetation and multiple soil properties for rangelands in the Northern Great Plains.  Specifically, we were interested in validating (or not) associations between measures of soil stability (e.g. water stable aggregates [WSA]) and two ecosystem properties (i.e. annual net primary productivity [ANPP] and water infiltration). 

We sampled a 0.3 ha area with equal sampling in a livestock exclosure (est. 1999) and two adjacent pastures.  Sampling included: ANPP, elevation, soil moisture, soil stability (i.e. WSA of two size classes of macroaggregates and rangeland health soil stability tests), and soil structure (i.e. field-saturated infiltrability and sorptivity). 

We used multiple linear regression (MLR) to determine that the greatest amount of variation in ANPP was explained by a model with field-saturated infiltrability (infiltration) and soil moisture (R²= 0.33).  Infiltration explained most of the variation in ANPP.  MLR was then used to identify the independent variables that best explained the variability in infiltration.  Moderate amounts of variation in infiltration were explained by elevation, sorptivity, and % WSA (0.25-1 mm size class) (R²= 0.39).  Evaluation of the squared semi-partial correlation coefficients revealed that elevation contributed the most and % WSA the least to explaining variation in infiltration. 

These results indicate that a measure of infiltration was the best overall predictor of ANPP.  These findings support the belief that plant growth is regulated by soil structure, though feedbacks between perennial plants and soil structure are likely.  Our findings failed to indicate that measures of soil stability were useful predictors of ANPP.  However, further analysis revealed that some measures of soil stability were useful predictors of water infiltration.  Our findings revealed only tenuous support for correlations between soil stability and measures of ecosystem health.  This has important implications for the validation of existing indicators of rangeland health (i.e. biotic integrity, hydrological function, and soil & site stability) which place considerable emphasis on measures of soil stability.  These results highlight the need for validation of metrics of rangeland/ecosystem health.