流域尺度根区蓄水能力估算及其对气象要素敏感性分析

根区蓄水能力(SR)在水文模拟、土壤水分运移以及植被生长发育等方面具有十分重要的作用。为对其进行流域尺度的估算,选取太子河上游南甸(峪)流域作为研究区,利用水量累积曲线法(mass curve technique,MCT)估算SR,基于流域出口水文站月径流深,对耦合了融雪模块的概念性水文模型FLEX进行参数率定,获取土壤蓄水能力Su Max,并与MCT的估算结果进行交叉验证。在此基础上设定不同气候情景,分析SR对降雨、融雪、蒸散发的敏感性。结果表明:1)改进的FLEX水文模型可用于模拟研究区径流,获得代表研究区土壤蓄水量多年平均水平的参数Su Max值为27 mm;2)水量累积曲线法得到的多年SR值服从耿贝尔(Gumbel)分布,且FLEX模型获得的根区蓄水能力检验了水量累积曲线法的估算结果;3)SR随蒸散发、降雨和融雪变化的曲线斜率分别为1.37、0.73和0.37,说明SR对蒸散发、降雨和融雪的敏感程度依次减弱,随着蒸散量增加、降雨量减少的幅度增大,敏感程度增强,融雪的增大或减少对SR的影响相近。研究结果为流域尺度根区缺水状况估算及维持植被系统生长发育提供一定依据。

Estimation of root zone storage capacity and sensitivity analysis to climate factors at catchment scales

Abstract: Moisture storage capacity at root zone plays an important role in hydrological processes, soil moisture movement and vegetation development, which is also a crucial parameter in hydrological and ecological modelling. However, due to the great heterogeneity in soil texture and the restriction of observation currently, there is no effective way to observe this parameter at catchment scales. In this study, a typical catchment of the Taizi River in Northeast China was selected as the study area. On the basis of monthly runoff depth data at Nandianyu Hydrological Station, the parameters in FLEX model were calibrated and validated during different time periods, of which the parameter Su represents the moisture storage capacity at root zone, while the SuMax corresponds to the Su under the best hydrological process condition in simulation. A snow model was incorporated in the original FLEX hydrological model in order to improve the performance of the model in the places where snow and melting water cannot be ignored. Based on observational meteorology data and multi-source remote sensing data sets, the modified mass curve technique (MCT) was employed to estimate the moisture storage capacity at root zone in Nandianyu catchment, taking snowmelt as an important part of water input for technique modification. The MCT was originally used in engineering design, however, in this study the moisture storage capacity was estimated using this approach by treating the whole root zone of this catchment as a reservoir. Based on the relationship between cumulative inflow and water demand in dry seasons when the rate of water demand exceeded water inflow, the required moisture storage capacity at root zone for each year was yielded, and cross validated with the result derived from the FLEX model to test the availability of MCT in the study area. Different climate scenarios were further set to test the sensitivity of moisture storage capacity at root zone to rainfall, snowmelt and evapotranspiration when these 3 climate factors increased or reduced by 10%, 20% and 30% independently using the MCT which demonstrated to be feasible. The results show that: 1) The improved FLEX model could be used to simulate hydrological process in the study area, presenting a high accuracy in runoff depth simulation and flow hydrograph simulation. The value of parameter SuMax representing the moisture storage capacity at root zone under the best simulating condition was 27 mm; 2) The moisture storage capacity at root zone estimated by MCT was demonstrated to follow the Gumbel distribution, with a range of 21-84 mm. The value of moisture storage capacity derived from this approach coincided with the estimate derived from the FLEX model, which meant that the modified MCT could be used to estimate the moisture storage capacity at root zone in the study area, and the estimates derived from this approach could be used as parameters in hydrological and ecological models; 3) Curve slopes of the change of moisture storage capacity at root zone with evapotranspiration, precipitation and snowmelt were 1.37, 0.73 and 0.37, showing that the moisture storage capacity at root zone had a reduced sensitivity to the change of evapotranspiration, precipitation and snowmelt. Besides, with the increase of evapotranspiration and reduction of precipitation, the sensitivity of the moisture storage capacity at root zone enhanced while there was no big difference under the variations of snowmelt. This work can provide a basis for water deficit estimation and maintaining normal development of ecosystems when they are faced with drought.