Modeling glacier melt and runoff in a high-altitude headwater catchment in the Cordillera Real, Andes
Modeling glacier melt and runoff in a high-altitude headwater catchment in the Cordillera Real, Andes
复制标题
对安第斯山脉雷亚尔山脉高海拔源头集水区的冰川融化和径流进行建模
DOI：
10.5194/hessd-10-13093-2013
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
Y. Asaoka
中科院分区：
文献类型：
--
作者：
T. Kinouchi;Tong Liu;J. Mendoza;Y. Asaoka
抽象的。部分冰川覆盖流域径流是冰川融化、融雪、冰川和非冰川地区融水和雨水的地表和地下径流的综合过程。此外，热带安第斯山脉的固有特征，如大的气象变化、高海拔和陡坡、湿地和湖泊的水文影响以及冰川快速退缩，使得难以模拟气候变化下的冰川水文响应。在这项研究中，我们开发了一个适用于热带安第斯山脉部分冰川流域的半分布式概念模型，考虑了所有这些方面，并将该模型应用于玻利维亚雷亚尔山脉的华纳波托西西源头流域。基于最新两年的气象水文监测数据，展示了该地区气温和降水的时空变化，并用于校准模型参数和验证日径流模拟的性能。模拟径流的变化与观测到的季节和时间变化非常吻合，结果还表明，气温和降水的时空变化的不确定性以及湿地和湖泊的阻滞效应强烈影响径流过程线。模拟径流成分表明，冰川融化径流主要发生在雨季初期（10月至12月上旬）和雨季后期（3月和4月），但后期径流量相对较小。在雨季的这两个时期之间，估计主要径流成分是非冰川地区的地下径流和融雪引起的地表径流。根据观测数据和预测性大气环流模型输出给出的未来气象条件，该模型量化了径流、冰川面积以及累积冰川和雪质量平衡的长期变化。冰川退缩预计将持续到 2050 年，面积减少的幅度和负累积质量平衡取决于所使用的温度升高趋势。特别是对于较高的温度趋势，径流的季节性变化更大，降雨对地下径流和地表径流的贡献更大，模拟出现在雨季，但从现在到2050年，年总径流的变化并不显着。这些结果表明，重要的是考虑如何最好地适应下游地区水资源可用性方面更大的季节性径流变化。
Abstract. Runoff from catchments with partial glacier cover is an integrated process of glacier melt, snowmelt, and surface and subsurface runoff of meltwater and rain from glacierized and non-glacierized areas. Additionally, inherent characteristics of the tropical Andes such as large meteorological variability, high elevation and steep slopes, hydrological effects of wetlands and lakes, and rapid glacier retreat make it difficult to model glacio-hydrological responses under changing climate. In this study, we developed a semi-distributed conceptual model applicable to partially glacierized catchments in the tropical Andes that considers all of these aspects, and we applied the model to the Huayna Potosi West headwater catchment in the Cordillera Real, Bolivia. Based on the latest 2 yr dataset of meteorological and hydrological monitoring, we showed the spatial and temporal variability of air temperature and precipitation in the region, and the dataset was used to calibrate model parameters and validate the performance of the daily runoff simulation. Variations in the simulated streamflow agreed well with the observed seasonal and temporal variations, and the result also showed that uncertainty pertaining to the spatial and temporal variations in air temperature and precipitation as well as the retarding effect of a wetland and lake strongly affected the runoff hydrograph. The simulated runoff components indicated that runoff from glacier melt occurs mainly in the initial period of the wet season, from October to early December, and in the late period of the wet season, March and April, although the runoff is relatively small in the latter period. Between these two periods in the wet season, major runoff components were estimated to be subsurface runoff in the non-glacierized area and surface runoff due to snowmelt. Given the future meteorological conditions based on the observational data and a predictive general circulation model output, the model quantified the long-term changes in runoff, glacierized area, and cumulative glacier and snow mass balance. The glacier retreat is estimated to continue to 2050, with the magnitude of area decrease and negative cumulative mass balance depending on the increasing temperature trend used. For higher temperature trends, in particular, greater seasonal variation in runoff and larger contributions from subsurface runoff and surface runoff by rainfall were simulated to occur in the wet season, but the change in annual total runoff between the present and 2050 was not significant. These results suggest that it is important to consider how to best adapt to greater seasonal runoff variations in terms of water availability in the downstream region.