In our latest Performance Report, we analyze BWI’s discharge forecast model’s performance for the La Tardoire – Lavaud catchment.
As climate change amplifies hydrological extremes, such as droughts and floods, regular monitoring of water resources is becoming increasingly important for optimal water resource management.
BWI employs a semi-distributed hydrological model. With the integration of machine learning techniques, it monitors and forecasts river discharges. This blog post delves into a detailed performance evaluation of BWI’s discharge prediction model. We specifically focus on the ‘La Tardoire’ river, with its outlet located at Lavaud.
The La Tardoire River emerges as a vital tributary of the Charente, enriching the departments of Haute-Vienne and Charente before ultimately joining the Charente River via its tributary, the Bonnieure. The Tardoire has a length of 114 km, emphasizing it’s status as one of the main tributaries of La Charente. It’s basin area is over 1000 square kilometers.
The departments of Charente and Charente-Maritime constitute the majority of the Charente river basin, situated in southwestern region of France. Adjacent to the Gironde and the Dordogne basins, the river flows through eye-catching landscapes, vineyards, and historical villages. The Nouère, its main principal tributary, stretches for about 50 kilometers before entering the Charente close to Angoulême. The Charente River’s 10,950 square kilometer catchment area stretches from the Massif Central, where it comes from, to Rochefort-sur-Mer, where it flows into the ocean.
Numerous smaller tributaries and streams can be found in the Charente basin, which contributes to it’s hydrology’s diversity. The Boutonne, Antenne, and Seugne are three of these tributaries that enhance the water network in the basin. Many lakes and reservoirs in the catchment supply crucial water supplies for business, agriculture, and leisure, boosting the area’s economic and environmental vibrancy.
BWI utilizes a robust model performance evaluation framework that considers a range of metrics. Some of these include Nash-Sutcliffe model efficiency scores (NSE), normalized NSE (NNSE), Kling-Gupta Efficiency (KGE), and their respective modified and normalized versions. The preferred metric within this framework is NNSE, which ranges between 0 and 1. A higher NNSE score indicates superior model performance. This metric is particularly useful for tracking the model’s performance over time and effectively communicating the results.
At the Lavaud outlet, the model achieves an NNSE of 0.839. This is a score that scientific publications classify as indicative of good model performance (NNSE > 0.65). The optimal parameter sets and the final state of the basin at the end of the training are recorded and utilized to initialize the subsequent training batch.
