Augmented Statistics for Hydroclimatic Extremes: Spearman, Mann–Kendall, and AI Classification for Drought Risk Mapping

Emanuela Genovese, Clemente Maesano and Vincenzo Barrile ()
Additional contact information
Emanuela Genovese: Department of Civil, Building and Environmental Engineering (DICEA), Sapienza University of Rome, Via Eudossiana, 18, 00184 Rome, Italy
Clemente Maesano: Department of Civil, Building and Environmental Engineering (DICEA), Sapienza University of Rome, Via Eudossiana, 18, 00184 Rome, Italy
Vincenzo Barrile: Department Of Civil Engineering, Energy, Environment and Materials (DICEAM), Mediterranea University of Reggio Calabria, Via Zehender, 89124 Reggio Calabria, Italy

Sustainability, 2025, vol. 17, issue 20, 1-18

Abstract: The effects of climate change are now evident on all scales, both global and local. Extreme events linked to climate change, such as heat islands and water bombs, are occurring with increasing frequency, causing significant harm to humans. Furthermore, rising temperatures also cause significant drought and desertification, which must be carefully assessed and analyzed. For this reason, with a view to evaluating environmentally sustainable development, the following research focuses on the variables that contribute to the reduction in local water availability in the province of Reggio Calabria, considering air temperature, evapotranspiration, precipitation, and available water resources. The Mann–Kendall test revealed a statistically significant increasing trend in air temperature (Z = +2.5, p < 0.01) and a decreasing tendency in precipitation, while the NDWI analysis indicated a reduction of about 34% in surface water resources between 2019 and 2023. The Spearman test showed strong negative correlations between temperature and water availability (ρ = −0.68) and between evapotranspiration and water availability (ρ = −0.66). Lastly, four artificial intelligence (AI) classifiers were compared: Random Forest, XGBoost, Gradient Boosting Decision Tree, and Logistic Regression. Random Forest performed the best, with 93% accuracy and 90% precision. The results confirm the strong negative dependence of temperature and evapotranspiration on water resources and identify Random Forest as the most reliable model for determining the area’s most at risk of drought.

Keywords: remote sensing; augmented statistics; statistics; climate change; drought (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2071-1050/17/20/9251/pdf (application/pdf)
https://www.mdpi.com/2071-1050/17/20/9251/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:17:y:2025:i:20:p:9251-:d:1774294

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().