The Spatial Concentration of Emerging Forms of Agro-Industrial Integration (Clusters) in the Samarkand Region, Uzbekistan
Abstract
Agro-industrial clusters have become an important instrument of agricultural transformation in Uzbekistan, yet their spatial organization remains insufficiently examined from a geographical perspective. This study investigates the spatial concentration, sectoral differentiation, and geographical characteristics associated with the distribution of agro-industrial clusters in the Samarkand Region. The analysis combines official statistical data, descriptive GIS-based spatial analysis, Location Quotient (LQ) assessment, Global Moran’s I, district-level correlation analysis, and exploratory scenario analysis. The results reveal pronounced territorial differentiation, although Global Moran’s I indicates no statistically significant region-wide spatial autocorrelation (I = −0.218, p = 0.346). Cotton–textile clusters are concentrated in districts with extensive irrigated agricultural land, with a significant positive association supported by both Pearson (r = 0.571, p = 0.033) and Spearman (ρ = 0.737, p = 0.003) correlations. Fruit and vegetable clusters also show descriptive concentration in irrigated districts, although no statistically significant district-level association was identified. Other cluster types exhibit spatial patterns associated descriptively with processing infrastructure, feed resources, market accessibility, and historical agricultural specialization. Illustrative scenarios suggest that future development may increasingly depend on production intensification, value-added processing, and export diversification under conditions of limited irrigated land and growing water scarcity. Conceptually, the findings show that Uzbekistan’s agro-industrial “clusters” function as state-initiated, vertically integrated production systems rather than classical market-driven Porterian clusters. Given the aggregated district-level and exploratory nature of the analysis, the findings should be interpreted as ecological spatial associations rather than causal or cluster-level mechanisms.