Evaluation of metric precision for a riparian forest survey.

This article evaluates the performance of a protocol to monitor riparian forests in western Oregon, United States based on the quality of the data obtained from a field survey. Precision is the criteria used to determine the quality of 19 field and 6 derived metrics. The derived metrics were calculated from the field data. The survey consisted of 110 riparian sites on public and private lands that were sampled during the summers of 1996 and 1997. In order to calculate metric precision, some of the field plots were re-measured. Metric precision was defined in terms of the coefficient of variability (CV) and standard deviation and then compared with a pre-defined data quality objective (DQO). A metric was considered precise if the CV met or exceeded the DQO. The geomorphology metrics were not precise while the forest stand inventory metrics and forest cover metrics, with some exceptions, were precise. The precision for many of the field and derived metrics compared favorably with the level of precision for similar metrics reported in the literature. Recommendations are made to improve the precision for some metrics and they include changing the way precision is calculated, re-defining the field protocol, or improving field training.