GDROM v2: An Inventory of Operation Variables and Rules for 2,017 Large Reservoirs across the CONUS

Effective reservoir operation is critical for achieving multiple water management objectives, while also representing a major human intervention in hydrologic processes. However, due to sparse or inconsistent observations of daily inflow, release, and storage, most existing models either simplify or omit reservoir behavior. To address this gap, we present GDROM v2, a nationwide dataset covering daily operation variable times series and derived operation rules for 2,017 reservoirs across the Contiguous United States (CONUS), building upon the original GDROM (Li et al., 2023, https://doi.org/10.4211/hs.63add4d5826a4b21a6546c571bdece10) of 452 reservoirs.
The dataset includes:
1. Time series of daily inflow, release, and storage (collected, cleaned, and normalized)
2. Operation rules for each reservoir expressed as “if–then–else” statements
3. Python scripts and instructions for running, training, and transferring GDROMs
GDROM v2 supports modeling, benchmarking, and policy evaluation related to large-scale reservoir management under variable data conditions. It provides both an empirical basis and a scalable framework for integrating reservoir operations into hydrologic and water resources studies.

高效水库调度对于实现多重水资源管理目标至关重要，同时也是水文过程中一项重要的人为干预活动。然而，由于日入库流量、出库流量及蓄水量的观测数据稀疏且不一致，现有多数模型要么简化水库调度行为，要么直接忽略该环节。为填补这一研究空白，我们推出GDROM v2：这是覆盖美国本土（Contiguous United States, CONUS）内2017座水库的全国性数据集，包含日调度变量时间序列与衍生调度规则，其基于包含452座水库的初代GDROM（Li等，2023，https://doi.org/10.4211/hs.63add4d5826a4b21a6546c571bdece10）构建而成。

该数据集包含以下内容：
1. 日入库流量、出库流量及蓄水量时间序列（已完成采集、清洗与归一化处理）
2. 各水库的调度规则，以"if–then–else"条件语句形式表达
3. 用于运行、训练与迁移GDROM的Python脚本及操作指南

GDROM v2可支撑可变数据条件下大规模水库调度相关的建模、基准测试与政策评估工作，为将水库调度纳入水文与水资源研究提供了实证基础与可扩展框架。