Spectrum of climate change and streamflow alteration at a watershed scale

Abstract

Worldwide, evidences of water cycle alteration and fresh water resources depletion are frequently reported with various magnitudes. This alteration in the hydrologic cycle is often regarded as a signal of the actual climate change. However, the debate on climate change seems to have preferentially focused on global-scale patterns such that the rich knowledge gathered in the domain is virtually less integrated to decision making at the watershed level. Indeed, the watershed apprehension of climate change is probably an imperative for sustainable water resources planning. The scope of the present study aligns with that imperative as it aims at conciliating patterns of climate change with observations of hydrologic alterations at the watershed level. Specifically, the paper describes the interplay between land-cover changes and the terrestrial water cycle disturbances under climate change at the global level. Thereafter, it reports a watershed-level analysis of streamflow, land-cover, PET and precipitation alteration. Specially, the case study focused on the Brazos River basin, located in the USA and shared by the states of Texas and New Mexico. From both regional and watershed prospects, signals of hydrologic alteration during the time period 1955–2014 are highlighted and then implications of climate change are discussed. The results show an overall longitudinal gradient of precipitation changes and a latitudinal gradient of PET changes across the Brazos watershed. However, these gradients of changes seem to be driven by regional climate components which extend beyond the physical boundary of the Brazos watershed. Mann–Kendall’s analysis of discharge time series (annual average, minimum and maximum) at 10 different stations exhibits meaningful contrasts from upstream to downstream. An assessment of land-cover changes shows critical patterns of landscape change across the watershed. The analyses depicted signals of urbanization sprawl and land-cover degradation. Specially, the significant statistical relationships observed between the time series of maximum green vegetation fraction (MGVF) and streamflow also indicate that the origin of the observed hydrologic alteration is anthropogenic. Ultimately, the results are discussed within the scope of climate change.