Abstract: Full coupling of physical processes, natural numerical coupling, and parsimonious but accurate data coupling are three key steps in efficient and accurate simulation of distributed hydrologic states in watersheds. Here we present a physically-based, spatially distributed hydrologic model (called PIHM) that utilizes all the three coupling strategies. Interception, snow melt, transpiration, evaporation, overland flow, subsurface flow, river flow, macropore based infiltration and lateral stormflow, as well as flow through and over hydraulic structures such as weirs and dams are some of the physical processes handled in the model. A semi-discrete, Finite-Volume approach is used to define the distributed process equations on discretized unit elements, in terms of a fully-coupled system of ordinary differential equations. An implicit Newton-Krylov based solver that utilizes adaptive time stepping provides a robust and stable solution. Data-coupling is aided by the use of constrained unstructured meshes, and a flexible data model incorporated within an open-source GIS tool (PIHMgis). The spatial adaptivity of the mesh elements and temporal adaptivity of the numerical solver facilitates capture of multiple spatio-temporal scales, allowing important insight into hydrologic process interactions. The implementation of the model has been performed for a mesoscale watershed in central PA (Little-Juniata Watershed, 845 km²). Model results are validated by comparison of observed and predicted streamflow and groundwater levels at multiple locations. The fully-coupled model unfolds a range of multiscale/multiprocess interactions including: 1) an apparent inverse relationship between fraction of total evapotranspiration rate due to transpiration and interception loss, 2) the role of forcing (precipitation, temperature and radiation), soil moisture and overland flow on evaporation-transpiration partitioning, 3) the importance of water table depth on evaporation-transpiration,  4) the influence of local upland topography and stream morphology on spatially distributed, asymmetric right-left bank river-aquifer interactions, and, 5) the role of macropore and topography on ground water recharge magnitude, time scale and spatial distribution.

Abstract: Distributed physical models for the space-time distribution of water, energy, vegetation, and mass flow require new strategies for data representation, model domain decomposition, a-priori parameterization, and visualization. The Geographic Information System (GIS) has been traditionally used to accomplish these data management functionalities in hydrologic applications. However, the interaction between the data management tools and the physical model are often loosely integrated and non-dynamic. This is because a) the data types, semantics, resolutions and formats for the physical model system and the distributed data or parameters may be different, with significant data preprocessing required before they can be shared, b) the management tools may not be accessible or shared by the GIS and physical model c) the individual systems may be operating system dependent or are driven by proprietary data structures. The impediment to seamless data flow between the two software components has the effect of increasing the model setup time and analysis time of model output results, and also makes it restrictive to perform sophisticated numerical modeling procedures (real time forecasting, sensitivity analysis etc.) that utilize extensive GIS data. These limitations can be offset to a large degree by developing an integrated software component that shares data between the (hydrologic) model and the GIS modules. We contend that the pre-requisite for the development of such an integrated software component is a “shared data-model” that is designed using an Object Oriented Strategy. Here we present the design of such a shared data model taking into consideration the data type descriptions, identification of dataclasses, relationships and constraints. The developed data model has been used as a method base for developing a coupled GIS interface to Penn State Integrated Hydrologic Model (PIHM) called PIHMgis..

Abstract: Distributed integrated hydrologic models are data and computationally intensive. In order to perform a high temporal and spatial resolution run of a large scale hydrologic model in feasible time, parallelized versions of hydrologic model can be run on a cluster of parallel processors.  An efficient implementation of such a parallelized hydrologic model requires proper partitioning of the model domain. This paper discusses and highlights several hydrologic, architectural and algorithmic issues which need to be incorporated in an efficient domain partitioning for parallel implementation of integrated distributed hydrologic model. Here we also compare a suite of partitioning algorithms, both geometry and graph theory based, in terms of their efficiency in a) minimizing interprocessor communication b) load balancing c) adaptability  to  constraints and e) capturing actual communication volume. Hybrid algorithms are found to be most effective in minimizing communication volume. But the performance of the algorithms gets adversely affected while trying to satisfy multiple architectural constraints. The algorithms have been implemented on unstructured decomposed domain of Great Salt Lake basin and are discussed vis-à-vis finite volume based Parallelized – Pennstate Integrated Hydrologic Model (P-PIHM).

Abstract: Identification of the spatial distribution of ephemeral, intermittent and perennial channel networks over large ungauged basins in semi-arid regions is an important step for hydrodynamic simulation, landscape evolution and riparian vegetation dynamics, and calibration of watershed models. This paper describes an automated method for detection of stream channels and their classification into ephemeral, intermittent and perennial reaches using ASTER Level 1B remote sensing data. The methodology involves calculation of normalized difference water index map of the area of interest followed by bi-level uni-modal thresholding to obtain the wet channel network. This step is repeated over a registered set of multi-temporal images followed by application of a change-detection algorithm to obtain a classified map of the channel network with reaches of different water retention time scales.

Abstract: It is generally accepted that the seasonal cycle of precipitation and temperature in cordillera of the western US exhibits a north–south pattern for annual, interannual and decadal time scales related to largescale climate patterns. In this paper we explore these relationships, with special attention to the role of local and regional physiographic, hydrogeologic and anthropogenic conditions on low-frequency climate and terrestrial response modes. The goal is to try to understand the spatio-temporal structure in historical precipitation, temperature and streamflow records (P–T–Q) in terms of climate, physiography, hydrogeology, and human impacts. Spatial coherence in time series is examined by classification of factor loadings from principal component analysis. Classification pattern of P–T–Q stations indicate that local physiography, the hydrogeology, and anthropogenic factors transform atmospheric forcing and terrestrial response into unique clusters. To study the temporal structure, dominant low-frequency oscillatory modes are identified for a region from historical P–T–Q records using singular spectrum analysis. Noise-free time trajectories are reconstructed from the extracted low-frequency modes (seasonal–decadal) for each contributing watershed area corresponding to streamflow observation stations, and the phase–plane plots are obtained. Together, the spatial classification and phase plane provides a means of detecting how large-scale hydroclimatic patterns relate to major landforms and anthropogenic impacts across the CRB. The main result of this paper is that resolving the relative impact of basin-wide patterns of climate, physiography and anthropogenic factors (irrigation, dams, etc.) on runoff response can be a useful tool for detection and attribution for each source of variability.