Urban hydrologic trend analysis based on rainfall and runoff data analysis and conceptual model calibration

• Published in: Hydrological processes Vol. 31; no. 6; pp. 1349 - 1359
• Main Authors: Sun, Siao, Barraud, Sylvie, Branger, Flora, Braud, Isabelle, Castebrunet, Hélène
• Format: Journal Article
• Language: English
• Published: Chichester Wiley Subscription Services, Inc 15-03-2017; Wiley
• Subjects: Calibration; Catchment area; Catchment areas; Catchments; Coefficients; Concentration time; conceptual rainfall‐runoff model; Data; Data analysis; Data processing; Detection; Earth Sciences; Environmental Engineering; Environmental Sciences; Flooding; Flow rates; Hydrologic analysis; Hydrologic data; Hydrologic processes; Hydrology; Mann‐Kendall test; Mathematical models; Parameter identification; Pollution abatement; Rain; Rainfall; Rainfall-runoff modeling; Rainfall-runoff relationships; Runoff; Runoff coefficient; Sciences of the Universe; Sensitivity analysis; Storm runoff; Storm sewers; Stormwater; Stormwater management; Trend analysis; Trends; Urban areas; Urban catchments; urban hydrology; Urban rainfall; Urban runoff; Urbanization; Water management; Water pollution; ANALYSE DE DONNEES; RELATION PLUIE DEBIT; data analysis; hydrology; URBANISATION; HYDROLOGIE URBAINE; urbanization; rainfall-runoff relationship
• ISSN: 0885-6087; 1099-1085
• DOI: 10.1002/hyp.11109

Urban stormwater is a major cause of urban flooding and natural water pollution. It is therefore important to assess any hydrologic trends in urban catchments for stormwater management and planning. This study addresses urban hydrological trend analysis by examining trends in variables that characterize hydrological processes. The original and modified Mann‐Kendall methods are applied to trend detection in two French catchments, that is, Chassieu and La Lechere, based on approximately 1 decade of data from local monitoring programs. In both catchments, no trend is found in the major hydrological process driver (i.e., rainfall variables), whereas increasing trends are detected in runoff flow rates. As a consequence, the runoff coefficients tend to increase during the study period, probably due to growing imperviousness with the local urbanization process. In addition, conceptual urban rainfall‐runoff model parameters, which are identified via model calibration with an event based approach, are examined. Trend detection results indicate that there is no trend in the time of concentration in Chassieu, whereas a decreasing trend is present in La Lechere, which, however, needs to be validated with additional data. Sensitivity analysis indicates that the original Mann‐Kendall method is not sensitive to a few noisy values in the data series.