Study on evapotranspiration in the Yarkand Oasis and its downstream riparian areas based on a high-resolution remote sensing model

Abstract

[Objective] The Yarkand Oasis, a vital oasis in northwestern China, confronts severe water scarcity and ecosystem vulnerability. High-spatial-resolution estimation of evapotranspiration(ET) is essential to understand water consumption patterns and support sustainable water resource allocation in this region. [Methods] A 30 m resolution remote sensing ET model was developed by integrating multi-source Landsat data, enabling refined ET estimation across the oasis and downstream riparian areas from 2000 to 2020. [Results] Model validation showed a relative deviation of approximately 10% compared to water balance estimations for total oasis ET. High ET values were concentrated in the irrigated lands and along riverbanks, with a peak daily average ET of approximately 6.1 mm/d. In contrast, low-value zones(daily ET <0.4 mm/d) dominated oasis margins and regions distant from river channels. Downstream riparian ET decreased progressively from the riverbanks outward, with higher ET within 2 km of the riverbanks, exceeding 1 mm/d in some areas. Since 2002, total water consumption in the downstream river channels has shown a fluctuating upward trend, with a multi-year average of 466 million m³, peaking in 2013 at approximately 880 million m³. Monthly mean ET displayed distinct seasonal variations, reaching a maximum of 1.0 mm/d in July. [Conclusion] The high-resolution model effectively captures fine-scale ET variations in sparsely vegetated areas, revealing significant spatiotemporal patterns of water consumption. These findings provide critical insights for water resource management and ecological restoration in arid oasis-desert systems.

Keywords

Yarkand Oasis / sparse vegetation / Landsat / remote sensing / precipitation / water resources / spatiotemporal distribution / evapotranspiration