A Policy-Oriented Spatial Evaluation Framework for Assessing Soil-Data Uncertainty in Agri-Environmental Indicators and Model-Derived Metrics: Greece as a Case Study

Agri-environmental policies increasingly rely on spatial indicators and model-derived metrics calculated from harmonized soil datasets, yet disagreement in the underlying soil information can propagate differently depending on the mathematical structure and decision rule of the derived output. This study develops a policy-oriented spatial evaluation framework for assessing dataset-induced soil-data uncertainty at the level of selected agri-environmental indicators and model-derived metrics, together with their potential policy interpretation, using Greece as a national-scale case study. Two agri-environmental indicators (the MEDALUS ESAI Soil Quality Index (SQI) and SOC/Clay ratio) and two model-derived metrics (the RUSLE soil-erodibility factor K and SCS Curve Number (CN)) were evaluated using selected JRC/ESDAC soil products, ISRIC SoilGrids, and an integrated Greek field-based reference dataset. Output-specific complete paired samples ranged from N = 12,049 for SQI to N = 16,573 for CN, with N = 15,066 for RUSLE K and N = 15,005 for SOC/Clay. The framework combined output-specific paired statistics with decision-relevant disagreement definitions, local spatial-clustering diagnostics, and terrain stratification. Results were strongly output-dependent. SQI showed high overall agreement, with overall accuracy of about 91% for both international datasets, but class-specific performance was substantially weaker for the highest-risk Class 3: Producer’s Accuracy was only about 54%, indicating that approximately 46% of Greek-reference Class 3 locations were assigned to lower-risk classes by the international datasets. RUSLE K was more sensitive: under the primary study-defined ±10% criterion, overestimation affected 80.76% of ESDAC and 62.86% of ISRIC locations, and this ordering persisted across the ±5% to ±15% sensitivity range. For SOC/Clay, threshold-crossing cases in which the Greek reference was at or below 1/13 but the international estimate was above 1/13 (“false-healthy” cases) occurred in 22.81% of ESDAC and 13.50% of ISRIC comparisons. CN had similar national central tendencies among datasets, yet policy-relevant underestimation affected approximately 34% of ESDAC and 27% of ISRIC locations. Terrain stratification revealed a clear SOC/Clay pattern: false-healthy threshold-crossing cases increased from 20.69% to 32.33% for ESDAC and from 10.49% to 26.20% for ISRIC between the <6% and 18–35% slope classes. Policy-relevant soil-data uncertainty was therefore not equivalent to numerical disagreement, and its propagation was strongly output-dependent. The SOC/Clay false-healthy result provides a clear example: a numerically higher estimate can produce a lower apparent policy concern. International gridded soil products remain valuable for broad-scale screening, but their fitness for purpose should be evaluated at the level of the derived output and the decision context in which it is used.

Authors

Institutions

Publication Details

Journal
Land
Published
2026-11-13
DOI
https://doi.org/10.3390/land15112177
Primary Topic
Soil Geostatistics and Mapping
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A Policy-Oriented Spatial Evaluation Framework for Assessing Soil-Data Uncertainty in Agri-Environmental Indicators and Model-Derived Metrics: Greece as a Case Study

Alexandros Stavropoulos, Dionissios P. Kalivas, Konstantinos X. Soulis, Stergia Palli Gravani et al.
Land
Soil Geostatistics and Mapping
article

A Policy-Oriented Spatial Evaluation Framework for Assessing Soil-Data Uncertainty in Agri-Environmental Indicators and Model-Derived Metrics: Greece as a Case Study

Alexandros Stavropoulos, Dionissios P. Kalivas, Konstantinos X. Soulis, Stergia Palli Gravani, Stylianos Gerontidis, Nikitakis Evangelos
article en

Abstract

Agri-environmental policies increasingly rely on spatial indicators and model-derived metrics calculated from harmonized soil datasets, yet disagreement in the underlying soil information can propagate differently depending on the mathematical structure and decision rule of the derived output. This study develops a policy-oriented spatial evaluation framework for assessing dataset-induced soil-data uncertainty at the level of selected agri-environmental indicators and model-derived metrics, together with their potential policy interpretation, using Greece as a national-scale case study. Two agri-environmental indicators (the MEDALUS ESAI Soil Quality Index (SQI) and SOC/Clay ratio) and two model-derived metrics (the RUSLE soil-erodibility factor K and SCS Curve Number (CN)) were evaluated using selected JRC/ESDAC soil products, ISRIC SoilGrids, and an integrated Greek field-based reference dataset. Output-specific complete paired samples ranged from N = 12,049 for SQI to N = 16,573 for CN, with N = 15,066 for RUSLE K and N = 15,005 for SOC/Clay. The framework combined output-specific paired statistics with decision-relevant disagreement definitions, local spatial-clustering diagnostics, and terrain stratification. Results were strongly output-dependent. SQI showed high overall agreement, with overall accuracy of about 91% for both international datasets, but class-specific performance was substantially weaker for the highest-risk Class 3: Producer’s Accuracy was only about 54%, indicating that approximately 46% of Greek-reference Class 3 locations were assigned to lower-risk classes by the international datasets. RUSLE K was more sensitive: under the primary study-defined ±10% criterion, overestimation affected 80.76% of ESDAC and 62.86% of ISRIC locations, and this ordering persisted across the ±5% to ±15% sensitivity range. For SOC/Clay, threshold-crossing cases in which the Greek reference was at or below 1/13 but the international estimate was above 1/13 (“false-healthy” cases) occurred in 22.81% of ESDAC and 13.50% of ISRIC comparisons. CN had similar national central tendencies among datasets, yet policy-relevant underestimation affected approximately 34% of ESDAC and 27% of ISRIC locations. Terrain stratification revealed a clear SOC/Clay pattern: false-healthy threshold-crossing cases increased from 20.69% to 32.33% for ESDAC and from 10.49% to 26.20% for ISRIC between the <6% and 18–35% slope classes. Policy-relevant soil-data uncertainty was therefore not equivalent to numerical disagreement, and its propagation was strongly output-dependent. The SOC/Clay false-healthy result provides a clear example: a numerically higher estimate can produce a lower apparent policy concern. International gridded soil products remain valuable for broad-scale screening, but their fitness for purpose should be evaluated at the level of the derived output and the decision context in which it is used.

LandVol. 15(11)
Agricultural University of Athens (GR)
European Commission
Life on land, Zero hunger
Soil Geostatistics and Mapping
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon