Investigating the Impacts of DEM Type, Resolution, and Noise on Extracted Hydro-Geomorphologic Parameters of Watersheds via GIS

Investigating the Impacts of DEM Type, Resolution, and Noise on Extracted Hydro-Geomorphologic Parameters of Watersheds via GIS

Vahid Nourani (University of Tabriz, Iran), Safa Mokhtarian Asl (University of Tabriz, Iran), Maryam Khosravi Sorkhkolaee (University of Tabriz, Iran), Aida Hosseini Baghanam (University of Tabriz, Iran) and Masoud Mehrvand (University of Stuttgart, Germany)
Copyright: © 2018 |Pages: 43
DOI: 10.4018/978-1-5225-5039-6.ch006

Abstract

Water resources management is dependent on knowledge and understanding of water quantity and quality information with the latest developments in information technology such as geographic information system (GIS) to develop effective hydrological modeling within the water-based systems. The efficiency of such hydrological modeling relies on the accuracy of applied data. In this way, the application of low-quality data in developing models for integrated management of water resources can impose irreparable financial and human resources and environmental costs in the catchment area. Thus, in regions that shortage of data is the issue, semi-distributed modeling is a useful tool. In this chapter, three aims are followed: (1) effect of digital elevation model (DEM) type and resolution on extracted hydro-geomorphologic parameters, (2) effect of wavelet-based de-noising method on extracted hydro-geomorphologic parameters, (3) determination of the optimal cell size to extract topographic attributes with good agreement to the real features.
Chapter Preview
Top

Introduction

Within the past few years, the analysis of factors affecting the hydrological processes within basins has been one of the most significant areas of research in the field of hydrologic modeling. One of the most important prerequisites of developing the map of watersheds and knowing their hydrologic characteristics is the knowledge of rainfall-runoff simulating at the outlets of basins or sub-basins which is one of the most significant challenges of hydrologists, especially in the least developed countries that suffer from the lack of adequate data required for the hydrologic modeling.

Hydrologic models have been generally classified into two categories: lumped and physically-based models. Lumped models act as a black-box model and estimate runoff only at the catchment outlet. These models cannot provide any information about the distribution of saturated areas within the basin; therefore, they are unable to describe how saturated areas distributed within the basin and what their role in evapotranspiration and runoff production is, such as Stanford IV (Crawford & Linsley, 1966) and ARNO (Todini, 1996) models.

Since physically-based models have been developed based on spatially distributed digital and remotely sensed data sets of features such as precipitation, elevation, vegetation, etc., the quality of applied data plays an important role in obtaining efficient models. The application of low quality data in developing models for integrated management of water resources can impose irreparable financial and human resources and environmental costs in the catchment area. To overcome this problem, the physically-based models are applied. Physically-based models that are divided into two categories (i.e., distributed and semi-distributed models) can be utilized in regions that shortage of data or inapplicability of high quality data is the issue (Chen et al., 2004). Due to the various parameters involved in the distributed models, they are considered completely variable on the watershed area, these models require huge data quantities of watershed characteristics to have proper hydrological models such as WetSpa (Water and Energy Transfer between Soil, Plants a) model Atmosphere (Wang et al., 1996). In order to obviate the complexity of distributed models, semi-distributed models are posed. Since the parameters of semi-distributed models have relative spatial changes in the watershed area, so these models need less data than distributed models in hydrological modeling. The semi-distributed models such as TOPMODEL (Topographic Model) that have less complexity in hydrological modeling are a good trade-off between the lumped models and the distributed models (Beven & Kirkby, 1979; Nourani & Mano, 2007). Semi-distributed hydrologic models are created according to hydro-geomorphologic characteristics of the watershed. Terrain analysis and extraction of hydro-geomorphologic characteristics are usually performed by employing the Digital Elevation Model (DEM), as a means of topography information representation of the watershed via software of Geographic Information System (GIS). Indeed GIS is a system designed to capture, store, manipulate, analyze, manage, and present all types of spatial data. GIS is a relatively broad term that can be applied in various branches of science and technology. It is undeniable that the emergence of GIS in hydrology has improved hydrological models to a great extent since the accessibility to spatial elements such as slope, aspect and watershed or catchment area, etc., became easier.

Many hydrological models utilize GIS to edit the input data and to display output data. According to Maidment (1993) GIS applications in hydrology are classified into four categories:

  • 1.

    Hydrological assessment and classification; GIS is utilized to extract the overall data of watershed, while the manually extraction of these data is so difficult.

  • 2.

    Determination the hydrological parameters; commonly, GIS is applied to extract the parameters of hydrological models, especially in physically-based models.

  • 3.

    Linking the hydrological models with GIS.

  • 4.

    Integrated hydrological models in GIS.

Complete Chapter List

Search this Book:
Reset