Читать книгу Approaches to Soil Health Analysis, Volume 1 - Группа авторов - Страница 14

Once a suite of soil health indicators has been selected, standard methods for collecting and handling samples in the field, processing them in the laboratory, analyzing them, and interpreting the data are needed for monitoring and making appropriate comparisons (Doran & Parkin, 1994). This is especially true for biological assays which can be more sensitive to how soil samples are collected and processed prior to analysis than to subtle differences in the analytical methods themselves. Currently, soil health measurement protocols vary widely and can therefore lead to inconsistent results and slow progress toward widely validated interpretation. This challenge is best addressed by standardization of a minimum dataset of methods used across organizations that collaborate nationally to make progress on interpretation and science‐based management recommendations (USDA‐NRCS, 2019b). Thus, ongoing efforts among public‐sector and commercial laboratories are needed to ensure preanalytical soil processing (i.e., degree of aggregation, sieving, grinding, etc.) and analytical methods are standardized. As with all soil chemical measurements (e.g., pH, salinity, extractable N, phosphorus, and potassium), biological and physical indicators generally have large spatial and temporal variation. Care thus needs to be taken not only with sampling (i.e., compositing enough subsamples to make inferences about a sampled area) but also sampling methods (soil volume and depth), timing of collection (seasonal or annual), and the statistical methods used for interpretation.

Volume 2 is also intended to help reduce analytical variation in the measurement of soil health indicators. This is important because, as previously shown by the standardization of NRCS inherent soil property characterization methods, standardization makes large‐scale data integration and comparisons feasible. Without rigorous standardization of soil health methods, variation among laboratories will hinder evaluation of changes over time and space and development of interpretations for various soil types and climate scenarios. This will in turn make regional and national compilations of soil health data very difficult to interpret.

Standardization of methods and protocols, along with appropriate proficiency testing, will facilitate collection of high‐quality data with a high degree of interpretability, which is needed to facilitate development and use of regionally‐appropriate interpretation functions (i.e., scoring algorithms). Those algorithms are needed to transform raw laboratory data into unitless (0 to 1) values that shows how well a specific soil is performing a production or environmental function. Such ratings can then be used for on farm management decision making. Private and public soil testing laboratories that use broadly standardized methods will therefore have the advantage of being able to offer broadly validated soil health testing and interpretation using functions and recommendations developed from a large dataset achieved through multiorganization public‐private partnership contributions.