<p>Rapid human-induced changes, such as climate change, population growth and rapid urbanization, are putting enormous stress on water resources. An accurate estimate of available water resources is a prerequisite for sustainable water resources planning and management. For gauged basins, historical records of hydrological observations are available, but for ungauged basins, the assessment of water availability is a challenging task. Therefore, the major focus of studies in ungauged basins is the development of appropriate tools that can accurately quantify hydrologic responses under various land use and climatic conditions. The reduction of the number of unknown parameters to be estimated is a key aspect in the development of hydrological models for ungauged basins.</p><p>This work is part of these issues and proposes an approach to reduce the complexity of hydrological models that include substantial uncertainties about the input data, initial and boundary conditions, model structure and parameters, owing to lack of data (i.e. for ungauged basins) and poor knowledge of hydrological response mechanisms. The case study of a basin of the District of Licola, located in the territory of the municipality of Giugliano, a city near Naples (southern Italy) is analyzed. Originally devoted to agriculture and grazing, it has been affected in the last decades by intense urbanization, which caused an increase in the impermeability of the soil cover. The increase in residential, commercial and production buildings has changed the functioning of the drainage network canals, compared to the original conditions, causing an increase in the frequency of flooding in the area. The semi-distributed hydrological model SWMM is adopted, which allows the subdivision of the basin in sub-basins according to land use and soil data.</p><p>Sensitivity Analysis (SA) is an effective approach to model simplification, providing an assessment of how much each input / parameter contributes to the output uncertainty. In general, SA is an essential part of model development, reducing uncertainties that have negative effects on the accuracy and reliability of simulated results. Specifically, in this study the SA is carried out with a method based on the decomposition of the variance of the peak flow and runoff volume, to quantitatively evaluate the contributions of single uncertain inputs/parameters that characterize the surface runoff with respect to different rainfall events, for both pervious and impervious areas. To this aim, the Fourier Amplitude Sensitivity Test (FAST) is implemented. This method allows quantifying not only the &#8220;main effect&#8221; of variance, but also provides the Total Sensitivity Indices (TSI), defined as the sum of all the sensitivity indices for each parameter (including the effects of the interaction with other uncertain parameters).</p><p>The research objectives aims at: (i) increased understanding of the relationships between input and output variables in a complex hydrological system; (ii) reduction of model uncertainty, through the identification of input parameters mostly contributing to output variability and should therefore be the focus of sensitivity analysis; (iii) model simplification, fixing&#160; the values of input parameters that have little effect on the output, and identifying and removing redundant parts of the model structure.</p>
An Integrative Model for Soil Biogeochemistry and Methane Processes: I. Model Structure and Sensitivity Analysis
