Hydrological non-stationarity in the Colorado River (Argentina) after 50 years: Implications for legacy planning models

Résumé fourni par la source

The Colorado River (CR) is a snow-dominated hydrological system of critical importance for central-western Argentina, where inter-jurisdictional water management is governed by the "Unique Agreed Program" (UAP). This framework was originally based on modeling conducted by the Massachusetts Institute of Technology (MIT) (1969–1973), which assumed stationary hydrological conditions and associated water quality patterns. This study evaluates the validity of that assumption under current hydrological conditions by analyzing updated streamflow and historical electrical conductivity (EC) series at Buta Ranquil (BR), Casa de Piedra (CdP), and Paso Alsina (PA). Structural changes were assessed using standard homogeneity tests (SNHT, Buishand, and Pettitt). Results reveal consistent non-homogeneity across all series, with a coherent breakpoint around the 2010/11 hydrological year, indicating a transition to a persistent deficit regime. At the headwaters (BR), mean streamflow declined from 150.6 m 3 /s to 86.9 m 3 /s, falling well below the original MIT design flow (147 m 3 /s). Downstream, CdP experienced a substantial reduction in water availability, with mean outflows decreasing from 133.5 m 3 /s to 62.7 m 3 /s. This hydrological decline was accompanied by a basin-wide increase in EC, primarily associated with reduced dilution capacity. However, EC values at PA remained below the critical threshold of 1800 μS/cm established by the UAP, highlighting the buffering role of CdP and coordinated basin management. These findings demonstrate that the stationarity assumption underpinning the UAP is no longer valid under current conditions. The study provides empirical evidence of a basin-scale regime shift and underscores the need for adaptive, non-stationary water management strategies to ensure long-term system resilience.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.