Classification of natural flow regimes of non-perennial rivers and streams (NPRS) is an incipient field of research. NPRS represent approximately 70% of the total Mediterranean rivers and are expected to increase in the next decades as a result of climate change. Due to the ecological importance of NPRS and the need to improve national ecological assessment methods within the scope of the Water Framework Directive (WFD), this paper aims to classify the hydrological regime of 69 non-regulated streams, testing several hydrological indices related to the frequency, duration, timing, and rate of change in periods of flow cessation. Using daily flow records, a total of 315 indices were calculated and their relationships were examined with Principal Component Analysis (PCA) for different thresholds used to define zero-flow condition set at 0, 1, 2, and 5 l/s. Redundancy analysis identified five indices that better describe the patterns of hydrological variability in Mediterranean NPRS: number of days per year without flow, annual percentage of months without flow, mean daily of annual flows, coefficient of variation of Julian date of the annual start of zero flow and annual rise rate. Using these indices, a self-organizing map (SOM) was trained in order to categorize the NPRS into three groups with similar hydrological features. The results of this study provide a statistically-based hydrological classification of NPRS in Mediterranean environments. We expect that this classification will provide useful insights to water authorities to improve the assessments of the ecological status in this set of water bodies.

The effects of the discharge of wastewater treatment plants (WWTP) on the status of rivers have most commonly been focused on water quality. A very limited number of works have characterised the ability of treatment plants to modify flow patterns in the receiving rivers. This paper presents a methodology for the assessment of the hydrologic alteration caused by WWTP discharges, over a two-fold sequence. The first phase comprises the application of indicators derived from accessible data and informative of the capacity of treatment plants to produce significant flow alterations. The second phase, which may only be carried out when flow data in the receiving river is available, is based on the indicators of hydrologic alteration provided by the free software IAHRIS (6 indicators) and IHA (2 indicators), and on a new indicator proposed in this paper to obtain information of flow alteration at seasonal and monthly time scales. The procedure suggested in this work is applied to the Manzanares River (Central Spain), allowing the quantification of the flow alteration generated by the 12 WWTP which give service to Madrid city (3.8 million inhabitants): Large increases of annual water volumes (from 108 hm3 to 410 hm3); at a monthly scale (increase from 246% to 1516%); variability in flow decreases in wet years by up to 47% and increases in dry years by up to 380%; seasonal patterns is altered within an altered regime. Results of the analysis show: (i) the ability of the proposed methodology to characterise the modification of flow patterns due to WWTP discharges; (ii) the importance of assessing such changes when evaluating the environmental impact of treatment plants; (iii) the importance of designing preventive and mitigation measures which maintain the ecological integrity of river ecosystems in the receiving channels.