This paper discusses a novel way of generating sampling points of hydrologic features, specifically streams, irrigation network and inland wetlands, that could provide a promising measure of accuracy using combinations of traditional statistical sampling methods. Traditional statistical sampling techniques such as simple random sampling, systematic sampling, stratified sampling and disproportionate random sampling were all designed to generate points in an area where all the cells are classified and subjected to actual field validation. However, these sampling techniques are not applicable when generating points along linear features. This paper presents the Weighted Disproportionate Stratified Systematic Random Sampling (WDSSRS), a tool that combines the systematic and disproportionate stratified random sampling methods in generating points for accuracy computation. This tool makes use of a map series boundary shapefile covering around 27 by 27 kilometers at a scale of 1:50000, and the LiDAR-extracted hydrologic features shapefiles (e.g. wetland polygons and linear features of stream and irrigation network). Using the map sheet shapefile, a 10 x 10 grid is generated, and grid cells with water and non-water features are tagged accordingly. Cells with water features are checked for the presence of intersecting linear features, and the intersections are given higher weights in the selection of validation points. The grid cells with non-intersecting linear features are then evaluated and the remaining points are generated randomly along these features. For grid cells with nonwater features, the sample points are generated randomly.
Stream network delineation based on LiDAR-derived digital terrain model (DTM) may produce stream segments that are inexistent or incomplete because of limitations imposed by extraction procedure, terrain and data. The applicability of a common threshold value in defining streams such as those implemented through the D8 algorithm also remains in question because the threshold varies depending on the geomorphology of the area. Flat areas and improper hydrologic conditioning produce erratic stream network. To counteract these limitations, this study proposes a workflow that improves the stream network produced by the D8 algorithm. It incorporates user-defined channel initiation points as inputs to a tool developed to automatically trace the flow of water into the next actual stream segment. Spurious streams along digital dams and flat areas are also manually reshaped. The proposed workflow is implemented in Iligan River Basin, Philippines using LiDARderived DTM of 1-meter resolution. The Flow Path Tracing (FPT) method counteracts the limits imposed by extraction procedure, terrain and data. It is applicable to different typologies of watersheds by eliminating the need to use site-specific threshold in determining streams. FPT is implemented as a Phyton script to automate the tracing of the streams using the flow direction raster. The FPT method is compared to the blue line digitization and the D8 method using morphometric parameters, such as stream number, stream order and stream length, to assess its performance. Results show that streams derived from the FPT method has higher stream order, number and length. An accuracy of 93.5% produced from field validation of the FPT method’s streams strengthens the findings that integrating manual channel head initiation and flow path tracing can be used for nationwide extraction of streams using LiDAR-derived-DTM in the Philippines.
Irrigation networks are important in distributing water resources to areas where rainfall is not enough to sustain agriculture. They are also crucial when it comes to being able to redirect vast amounts of water to decrease the risks of flooding in flat areas, especially near sources of water. With the lack of studies about irrigation feature extraction, which range from wide canals to small ditches, this study aims to present a method of extracting these features from LiDAR-derived digital terrain models (DTMs) using Geographic Information Systems (GIS) tools such as ArcGIS and Whitebox Geospatial Analysis Tools (Whitebox GAT). High-resolution LiDAR DTMs with 1-meter horizontal and 0.25-meter vertical accuracies were processed to generate the gully depth map. This map was then reclassified, converted to vector, and filtered according to segment length, and sinuosity to be able to isolate these irrigation features. Initial results in the test area show that the extraction completeness is greater than 80% when compared with data obtained from the National Irrigation Administration (NIA).
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.