URA dataset - 40 Basque Country catchments hourly hydro-meteorological data + Catchments Attributes

Basque Country is located in north of Spain on the Atlantic coast. This region, characterized by its humid climatology, has abundant water resources. The catchments, noted for their flashy and humid characteristics, span an area of 4494 Km2 and include a diverse range of basin sizes from 4 to 1000 Km2. The Basque Water Agency (URA), a regional governmental entity in the Basque Country, is responsible for managing water policies and resources in this territory. To support water resources planning and management, URA has compiled a high-quality dataset suitable for data mining using deep learning models. This dataset includes hourly hydro-meteorological timeseries, reflecting the region’s steep, flashy, and humid hydrological dynamics. URA catchments are situated between the Cantabrian Mountains (reaching up to 1300 meters in elevation) in the northwest and the Atlantic Ocean to the north. The region is predominantly covered by grasslands and evergreen forests and benefits from the warming effects of the Gulf Stream. The climate is humid and temperate, with mean annual temperatures ranging from 9°C in the mountains to 15°C in lower regions. Annual rainfall varies between 1200 and 1600 millimeters, primarily due to the advection of North Atlantic fronts. URA has collected hourly hydro-meteorological timeseries from approximately 100 stations distributed across the region, including rain gauges and water level measurement sites. Accurate rainfall-runoff modeling is critical in this region due to its susceptibility to flash floods. This dataset presents 21 years of hourly timeseries from 40 catchments within the region. This extensive temporal coverage and high-resolution data offer valuable insights into the region’s hydrological dynamics. This Dataset is published with the papers: 1- Hosseini, F.; Prieto, C.; Álvarez, C. (2025). Ensemble learning of catchment-wise optimized LSTMs enhances regional rainfall-runoff modelling—Case Study: Basque Country, Spain. J. Hydrol. 132269. doi: 10.1016/j.jhydrol.2024.132269 2- Hosseini, F., Prieto, C., & Álvarez, C. (2024a). Hyperparameter optimization of regional hydrological LSTMs by random search: A case study from Basque Country, Spain. Journal of Hydrology, 132003. doi: 10.1016/j.jhydrol.2024.132003巴斯克地区（Basque Country）位于西班牙北部大西洋沿岸地带。该区域以湿润气候为典型特征，水资源禀赋丰沛。本次研究涉及的流域以水文突发性强、气候湿润为显著特点，总面积达4494平方千米，涵盖4至1000平方千米不等的多样流域尺度。 巴斯克水务署（Basque Water Agency，简称URA）是巴斯克地区的区级政府机构，负责统筹该辖区内的水政策制定与水资源管理工作。为支撑水资源规划与管理业务，巴斯克水务署编制了一套适用于深度学习模型数据挖掘的高质量数据集。该数据集包含逐小时水文气象时间序列，精准反映了该区域地形陡峭、水文突发性强且气候湿润的动态特征。 该数据集覆盖的巴斯克流域位于西北部坎塔布连山脉（Cantabrian Mountains，最高海拔达1300米）与北部大西洋之间。区域地表覆被以草原与常绿阔叶林为主，受墨西哥湾暖流（Gulf Stream）的增温效应加持。气候整体湿润温和，年平均气温由山地的9℃至低海拔区域的15℃不等；年降雨量介于1200至1600毫米之间，降水主要源自北大西洋锋面的平流输送。 巴斯克水务署通过分布在全区的约100个监测站点（涵盖雨量站与水位测量站点）采集逐小时水文气象时间序列数据。鉴于该区域极易发生山洪灾害，精准的降雨径流模拟工作至关重要。本数据集包含该区域内40个流域的21年逐小时时间序列数据，其长时间序列覆盖范围与高分辨率特性，为深入解析区域水文动态提供了宝贵的研究基础。 本数据集已随以下两篇学术论文正式发表： 1. Hosseini, F.; Prieto, C.; Álvarez, C. (2025). 面向流域优化的长短期记忆网络（Long Short-Term Memory，LSTM）集成学习方法提升区域降雨径流模拟性能——以西班牙巴斯克地区为例. 《水文学报》, 132269. DOI: 10.1016/j.jhydrol.2024.132269 2. Hosseini, F., Prieto, C., & Álvarez, C. (2024a). 基于随机搜索的区域水文LSTM超参数优化——以西班牙巴斯克地区为例. 《水文学报》, 132003. DOI: 10.1016/j.jhydrol.2024.132003