Identifying and quantifying the latent factors controlling spring discharge in an industrialized karst watershed

Abstract

Karst springs have largely been impacted by human activity and climate change, which increases the uncertainties to supply and management of water resources. It’s essential to understand the footprint of the controlling factors on their hydrological behavior. Taking the example of Niangziguan springs in China, this study explores the time series of spring discharge over a 50-year time-window (1959–2007). Local Hurst exponent is employed to analyze its temporal correlation behavior. The prime latent factors of spring discharge are extracted and identified using dynamic factor analysis and wavelet analysis, respectively. Moreover, we quantified the contributions of latent factors to spring discharge. The results exhibit that the spring discharge is controlled by the positive feedback mechanism before the exploitation of karst groundwater, however, human activities lead to the feature of noise for spring discharge fluctuations. With the development of industry and agriculture, the effect of the combination of controlling factors on spring discharge tends to be stable. Before groundwater exploited, the local recharge cycle of groundwater shows the largest contribution (43%) on spring discharge focusing on the scales of 4–16 months, and the construction of the industrial region weakens the impact of local groundwater recharge (41%). The contribution of human activities, regional groundwater recharge, and rainfall/runoff to spring discharge (27%-30%) is almost equal. Therefore, human activities have become one of the most important factors controlling the spring discharge. The WNPM (West North Pacific Monsoon) and ISM (Indian Summer Monsoon) show the smallest effect (3%) on spring discharge.