A topsoil classification framework for soil health using k ‐means clustering applied to Florida soils
Soil Science Society of America Journal · 2026-11 · 90 (6)
DOI: 10.1002/saj2.70408 ↗Kurz gefasst
Fachartikel in der Zeitschrift Soil Science Society of America Journal, online erschienen am 17.11.2026. Verfasst von Yaslin N. Gonzalez und 2 weiteren Autor:innen. Die Arbeit nennt 95 Literaturangaben. Gefördert von: National Institute of Food and Agriculture.
Abstract (Originalsprache)
Soil health assessments measure dynamic soil properties to evaluate management impacts on soil functions, such as nutrient cycling and water storage; however inherent soil properties (e.g., texture) can often exert stronger influences. One approach to capture regional differences in inherent and (semi‐)dynamic topsoil properties is to use the k ‐means algorithm to identify soil health regions using the gridded Soil Survey Geographic database. Using this framework, we grouped Florida surface (0‐ to 30‐cm depth) soils into eight regions, validated using clustering metrics. Scoring curves were developed by region and statewide for soil organic matter (SOM), saturated hydraulic conductivity, available water capacity (AWC), bulk density (BD), electrical conductivity, and soil pH. Most regions are characterized by coarse‐textured soils (15–57 g kg −1 clay) that are acidic (pH 5.2–5.9) and low in SOM (6–19 g kg −1 ). However, SOM and clay content are higher in Regions 2, 6, and 8, consisting primarily of wetlands. Region 2 achieved the highest overall soil health score (0.70), while all other regions received medium scores (0.41–0.51), suggesting widespread inherent limitations in soil functions. For instance, Region 3—the most expansive area—contains most cropland, yet productivity may be constrained by low AWC, high BD, and low SOM. Ultimately, this framework provides regional average conditions from which on‐site measurements can be compared to interpret management‐induced changes.
Förderung
- National Institute of Food and Agriculture
Zitieren
APA
Gonzalez, Y. N., Chatterjee, A., & Maltais‐Landry, G. (2026). A topsoil classification framework for soil health using k ‐means clustering applied to Florida soils. Soil Science Society of America Journal, 90(6). https://doi.org/10.1002/saj2.70408
Vancouver
Gonzalez YN, Chatterjee A, Maltais‐Landry G. A topsoil classification framework for soil health using k ‐means clustering applied to Florida soils. Soil Science Society of America Journal. 2026;90(6). doi:10.1002/saj2.70408
BibTeX
@article{Gonzalez2026,
author = {Gonzalez, Yaslin N. and Chatterjee, Amitava and Maltais‐Landry, Gabriel},
title = {A topsoil classification framework for soil health using k ‐means clustering applied to Florida soils},
journal = {Soil Science Society of America Journal},
year = {2026},
volume = {90},
number = {6},
doi = {10.1002/saj2.70408}
}
EUVIMED