梯级水电站运行条件下长江上游鲢早期资源时空分布格局

Spatial and temporal distribution of Hypophthalmichthys molitrix early-stage resources in the upper reaches of the Yangtze River under the operation condition of cascade hydropower station

为探究在金沙江梯级水电站运行的影响下,长江上游鲢(Hypophthalmichthys molitrix)早期资源分布及补充机制,选取2017—2019年5—7月长江上游早期资源调查数据,并运用基于Tweedie分布的广义相加模型(GAM)分析影响其时空分布格局的主要环境因子。结果表明,2017—2019年鱼卵密度呈现一致的变化趋势。在空间上,距离大坝最远,流量、水温最高,透明度较低的江津江段鲢鱼卵密度最高,宜宾江段未调查到鲢鱼卵;时间上,6月的鱼卵密度显著高于5月和7月,推测6月为长江上游鲢的繁殖盛期。广义相加模型结果显示,显著影响鲢鱼卵密度空间分布的主要环境因子为流量、离坝距离、透明度和水温(P<0.05)。在水温21~23℃,流量6000~14000m^(3)/s,透明度小于50 cm的范围内,鲢鱼卵密度分布较为集中。综合上述结果表明,金沙江梯级水电站运行下泄的低温、低流量和高透明度水严重影响了大坝附近江段的鲢自然繁殖,而随着与大坝距离的增加和沿江支流的汇入,大坝对鱼类繁殖的影响程度逐渐削弱,从而形成纵向恢复梯度。产卵量与水文指标相关分析表明,产卵量与涨水持续时间和流量涨幅呈正相关(P<0.05)。当涨水时间持续4 d及以上,流量累积涨幅为5030 m^(3)/s及以上时,有利于长江上游鲢自然繁殖。此外,研究发现,三峡大坝建成后,鲢产卵活动呈现向上游迁移的趋势,长江上游白沙镇至榕山镇江段为鲢的主要产卵场。因此,建议将距坝较远的江段,特别是库尾以上重庆至泸州江段作为重点繁殖区域加以保护,利于鲢补充群体的维持和壮大。

Dam construction alters natural river flows and hydrological regimes,including water temperature regimes,and this is a key environmental factor influencing fish reproduction.The aim of this study was to investigate the distribution and replenishment mechanism of the early population of silver cap,Hypophthalmichthys molitrix under the influence of the cascade hydropower station on Jinsha River Dam.Based on the early resource survey data of three sections[Yibin(YB),Luzhou(LZ)and Jiangjin(JJ)]in the upper Yangtze River reaches taken from May to July of 2017 to 2019,generalized additive models with Tweedie distribution were used to analyze the major environmental factors influencing their temporal and spatial patterns.The results revealed that the density of silver carp eggs showed a consistent trend from 2017 to 2019.Spatially,no H.molitrix egg was collected at YB,which was closest to the dam in the three river sections,whereas the density of H.molitrix eggs was highest at JJ,which was the farthest from the dam with the highest flow and water temperature and the lowest water transparency.Temporally,the density of fish eggs in June was significantly higher than that in May and July,suggesting that June was the peak breeding period of H.molitrix in the upper reaches of the Yangtze River.Generalized additive model analysis indicated that water discharge,distance from the dam,water transparency,and water temperature were the main environmental factors significantly affecting the spatial distribution of fish egg density with different relative contributions,together explaining more than 48.9%of total deviance observed(P<0.05).The highest interpretation rate of the deviation of the flow was 29.4%,followed by 8.7%by the distance from the dam.Density distribution of H.molitrix egg were mainly in the range of water temperature 21-23℃,flow rate 6000-14000 m^(3)/s,and transparency of less than 50 cm.Comprehensive analysis indicates that the low-temperature,low-flow,and high-transparency water released by the dam operation impact the natural reproduction of H.molitrix in the river section near the dam,while the degree of dam impact on fish reproduction is gradually weakened by increasing distance on the dam and the inflow of tributaries along the river,thus forming the longitudinal recovery gradients.The correlation analysis between spawning yield and hydrological indexes showed that the quantity of fish eggs was positively correlated with the duration of water rise and the increase of discharge(P<0.05).The natural reproduction of H.molitrix in the upper reaches of the Yangtze River benefits when the duration of water rise lasts for more than four days and the cumulative increase of flow exceeds 5030 m^(3)/s.Additionally,the spawning activities of H.molitrix showed a tendency to migrate upstream in the post-dam of the Three Gorges Dam,and the upper reaches of the Yangtze River from Baisha Town to Rongshan Town were the main spawning grounds of H.molitrix.Hence,the river sections further away from the dam,especially the Chongqing-Luzhou section above the reservoir tail,are suggested to be protected as the key breeding area for the growth and sustenance of H.molitrix population.