Abstract: Freshwater fish diversity is a key indicator of river ecosystem health,and studying its spatial patterns and underlying driving factors is essential for watershed ecological conservation. Traditional monitoring methods are limited by low efficiency and significant disturbance, whereas environmental DNA(eDNA) technology offers a new alternative for efficiently assessing fish diversity. This study investigated fish community composition,diversity patterns,and their environmental drivers in the Shaanxi section of the Weihe River-the largest tributary of the Yellow River-using eDNA metabarcoding. eDNA water samples were collected from 40 sampling sites during the normal water period in September 2023 and the dry period in April 2024,followed by high-throughput sequencing analysis. The results identified a total of 51 fish species belonging to 5 orders,14 families,and 38 genera,with 40 species recorded in the normal water period and 24 species in the dry period. Cypriniformes dominated in both periods,with Cyprinidae and Cobitidae families being predominant.Pseudorasbora parva,Zacco platypus,Hemiculter leucisculus,Rhodeus lighti,and Odontobutis potamophila were the shared dominant species across both periods. No significant differences in α diversity indices were found between the two periods, but community structure varied significantly, with sampling sites during normal water period exhibiting stronger heterogeneity and higher within-group β diversity. Redundancy analysis identified precipitation as the primary environmental factor affecting fish distribution in the Shaanxi section of the Weihe River Basin,followed by water temperature, five-day biochemical oxygen demand, and altitude. These findings clarify the response characteristics of fish communities to seasonal hydrological variation in the Shaanxi section of the Weihe River Basin and provide a scientific basis for regional aquatic ecosystem conservation and fish diversity management.