The expansion of planting area dominates the long term changes of water footprint for cotton in Xinjiang

Keywords: water footprint; cotton; spatial-temporal dynamics; climate change; xinjiang

Introduction: Cotton stands as a cash crop with high water consumption, yielding necessary benefits for human beings. However, more that 90% of cotton productivity in China are cultivated in Xinjiang, a water scare region, challenged the sustainability of agricultural development. Efforts on supply and demand of water resource for cotton in such region is critical for sustainable water management, but remains unresolved.

Material and Methods: Taking cotton cultivated in Xinjiang from 2001 to 2020 over counties as a case, we employed water footprint concept that based on virtual water to depict the spatial-temporal trend of water footprint, spatial clustering patterns of water footprint over county. We further identified the contributions of climate change, planting area, and inputs of fertilizer application to the changes of water footprint over regions.

Results: From 2001 to 2020, the average annual water footprint of cotton production in Xinjiang was 9.75 Gm³, with blue, green, and grey water footprints contributes 6.78 Gm³, 1.01 Gm³, and 1.96 Gm³, respectively. The overall water footprint exhibited an initial increase followed by a subsequent decrease, reaching its peak in 2014. Notably, the distribution of water footprints associated with cotton production varied across the study regions, with the average annual water footprints for cotton production in Southern Xinjiang and Northern Xinjiang recorded at 6.91 Gm³ and 2.84 Gm³, respectively. Over the study period, primary concentrations of the total water footprint of cotton were observed in the southern Tianshan Mountains, with no significant shifts in spatial aggregation at the county scale. The expansion of cotton cultivation areas and excessive fertilizer applications emerged as the main factors influencing the long-term dynamics of the cotton water footprint contributing 8.30 Gm³ and 1.26 Gm³ respectively, to the overall water footprint variation. Furthermore, climate change led to a reduction of 0.85 Gm³ in the water footprint of cotton production. The water footprint per unit yield of cotton within the study area exhibited a declining trajectory over the past two decades, with the average annual water footprint per unit yield calculated at 4845.91 m³/t.

Conclusions: the expansion of the planting area emerges as the primary driving force behind the dynamic shifts in the water footprint of cotton production in Xinjiang. Despite the overall increase in total cotton production, there is a notable downward trend in the water footprint per unit yield of cotton. This study provides a theoretical basis for balancing the sustainability of water use and the optimization of spatial patterns of cotton.