A Response to “Reanalysis Validates Soil Health Indicator Sensitivity and Correlation with Long‐term Crop Yields”

Abstract

A Response to "Reanalysis Validates Soil Health Indicator Sensitivity and Correlation with Long-term Crop Yields" Comment We published data showing that current soil health indicator (SHI) assessments do not consistently detect differences in a range of soil management practices implemented in North Carolina soils. Van Es and Karlen reanalyzed our data and asserted that it validates SHI correlation to crop yields and sensitivity to management as measured by the Comprehensive Assessment of Soil Health (CASH). We respond to van Es and Karlen with a more representative analysis of our data showing that individual SHI measurements are not predictive of crop yield from the 30-yr North Carolina agronomic trial. Regressions for aggregate stability (r 2 = 0.07) and P (r 2 = 0.18) show that neither SHI sufficiently predicts corn yield for this dataset and show no obvious pattern based on tillage intensity. Relationships between corn (Zea mays L.) yield and most biological SHI had r 2 £ 0.18, with only soil protein being moderately predictive of corn yield (r 2 = 0.45). The CASH index to assess overall soil health by integrating physical, chemical, and biological SHI measurements into a single value of soil health is also not predictive of corn yield in the trial (r 2 = 0.12). It is possible that current sampling and analytical procedures for assessing soil health do not consistently detect differences in productivity from soils with regional differences in land and ecological resources. We believe that calibrating SHI assessments to quantifiable agroecological outcomes instead of statistical rankings will reduce bias across regions and create a more inclusive framework for quantifying soil health.

Relationships of on‐farm soil health scores with corn and soybean yield in the midwestern United States

“…Two advantages of the CASH include scoring simplicity and a calibration dataset derived from many on-farm soil samples. However, a challenge to a broader application has been that a large proportion of the soil samples in the original calibration dataset came from the northeastern and midwestern United States (Fine et al, 2017;Roper et al, 2019;Stott, 2019); however, the sample database is expanding.…”

“…Moreover, despite several studies investigating how overall (or composite) soil health is accounted for by a single or a select few indicators (Fine et al, 2017;Frost et al, 2019;Rekik et al, 2018), these studies did not evaluate whether selected indicators represent composite soil health differently according to soil texture. Furthermore, several studies (Garland et al, 2021;Roper et al, 2019;Sainju et al, 2021;Wade et al, 2020;Wu & Congreves, 2021) have examined correlations among crop yields and soil health indicator observed values or assessment scores. However, these studies typically use historical yields and a single or few years of soil health assessment data rather than year-over-year yield and soil health data from many locations.…”

Sensitivity and variability of soil health indicators in a California cropping system

“…Therefore, how best to assess whether soil health is actually improving is a topic of great practical importance and much recent debate (e.g. Karlen, Veum, Sudduth, Obryki & Nunes., 2019;Norris et al, 2020;Roper, Osmond & Heitman, 2019;Stewart et al, 2018, Wander et al, 2019.…”

“…Accordingly, "soil health building practices" are increasingly being experimented with by farmers and incentivized by government and private entities. Therefore, how best to assess whether soil health is actually improving is a topic of great practical importance and much recent debate (Karlen et al, 2019;Norris et al, 2020;Roper et al, 2019;Stewart et al, 2018;Wander et al, 2019). Central to this debate are the concepts of indicator sensitivity, variability, and generalizability.…”

Base cation saturation ratios, soil health, and yield in organic field crops

“…These biologically mediated processes can be inferred through indicators of soil biological health, such as permanganate oxidizable C (POXC) (Weil et al, 2003), mineralizable C (Franzluebbers et al, 2000), and soil protein (Hurriso, Moebius-Clune, et al, 2018). These indicators represent the active and rapidly cycled pool of soil organic matter (SOM) and are often considered robust soil biological health indicators that are sensitive to changes in management (Culman et al, 2012;Hurisso et al, 2016;Roper et al, 2019).…”