Vegetation and water resource variability within the Köppen Geiger global climate classification scheme: a probabilistic interpretation

Authors: Sohoulande, Clement

Submitted to: Journal of Theoretical and Applied Climatology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/24/2023
Publication Date: 10/5/2023
Citation: Sohoulande Djebou, D.C. 2023. Vegetation and water resource variability within the Köppen Geiger global climate classification scheme: a probabilistic interpretation. Journal of Theoretical and Applied Climatology. 155:1081–1092. https://doi.org/10.1007/s00704-023-04682-z.
DOI: https://doi.org/10.1007/s00704-023-04682-z

Interpretive Summary: Worldwide, climate classifications are useful for characterizing ecosystems. Among these classifications, the Köppen-Geiger climate classification system has been increasingly used in a wide range of environmental fields across the globe. However, the Köppen-Geiger climate information is not sufficient to fully comprehend changes in vegetation and water resources. In fact, the spatial distribution of vegetation and water resources is known to be highly influenced by the variability of bioclimatic factors such as precipitation and temperature. Hence, connecting global variability of these bioclimatic factors to the Köppen-Geiger classification system could leverage the interpretability of climate classes. This study developed a quantitative approach for characterizing the variability within the Köppen-Geiger climate classification system. The bioclimatic variables used are precipitation, surface temperature, leaf area index, and liquid water equivalence anomalies. In the approach, index values were quantified and combined with models to estimate Köppen-Geiger climate classes. The study outcomes provide a better understanding of the variability within the Köppen-Geiger classification system and can therefore be used as a tool for interpreting climate variability.

Technical Abstract: Worldwide, climate classification schemes are useful for characterizing bioclimatic potentials of terrestrial ecosystems. Among these schemes, the Köppen-Geiger climate classification system has been increasingly used in a wide range of environmental fields across the globe. However, the Köppen-Geiger climate information is not sufficient to fully comprehend the dynamics of ecosystem components such as vegetation and water resources. In fact, the spatial distribution of vegetation and water resources is known to be highly influenced by the variability of bioclimatic factors. Hence, connecting global bioclimatic variability to the Köppen-Geiger classification system could leverage the interpretability of climate classes at the regional level. This study developed a combined entropy and probabilistic approach for characterizing bioclimatic variability within the Köppen-Geiger climate classification system. The bioclimatic variables used are half-degree gridded precipitation, surface temperature, leaf area index, and liquid water equivalence anomalies. In the approach, entropy-based disorder index (DI) values were quantified for individual variables and thresholds of DI percentiles were used to discretize bioclimatic variability zones at the global scale. Multivariate logistic regression models were later applied to the DI zone attributes and long-term averages of bioclimatic variables to estimate Köppen-Geiger climate classes. Statistics sustain variable models’ likelihoods (0.04=McFadden’s pseudo R2=0.92) but consistent estimates (0.87=Count R2=0.99) within the global climate classes.