LiDar Intensity for Improved Dection of inundation below the Forest Canopy

Authors: Lang, Megan; McCarty, Gregory

Submitted to: Wetlands
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/15/2009
Publication Date: 12/15/2009
Citation: Lang, M.W., McCarty, G.W. 2009. LiDar intensity for improved detection of inundation below the forest canopy. Wetlands. 29:1166-1178.

Interpretive Summary: Historically, the value of wetlands was largely unappreciated. Wetlands within the United States have been and are continuing to disappear at a rapid rate. Efforts are being made to conserve remaining wetlands and many policies have been adopted in support of this goal. To regulate the loss, preservation, and/or restoration of wetlands and to judge the effectiveness of these efforts in preserving associated ecosystem services, wetlands must be routinely monitored. Wetland hydrology (e.g., flooding) is the most important non-biologic factor controlling where wetlands are found and what process occur in wetlands, and should therefore be a vital part of any wetland mapping or monitoring program. Watershed-scale forested wetland hydrology has been difficult to study with conventional remote sensing methods and the collection of this information on the ground is cost prohibitive. However, this information is critical to informing the wetland regulation debate (e.g., which wetlands are protected under the Clean Water Act) ongoing within the federal government. Airborne Light Detection and Ranging (LiDAR) is a new and rapidly developing technology. LiDARs are active sensors, sending and receiving their own energy. LiDARs send out short pulses of energy and these pulses illuminate very small portions of the land’s surface. LiDAR data can be used to calculate precise locations (position and elevation). LiDAR data have mainly been used to derive information on elevation. However, the intensity (strength) of the signal used to derive information on elevation has the potential to significantly improve the ability to remotely monitor forested wetland hydrology. We demonstrate the ability to map wetland hydrology with greater than 96% accuracy using LiDAR intensity data. This method was found to be substantially more accurate than what could be obtained using traditional aerial photography. Our work demonstrated that LiDAR intensity data can be a key technology for addressing the regulatory status of wetlands and measuring the delivery of ecosystem services (e.g., flood water mitigation and the reduction of pollutants).

Technical Abstract: To regulate the loss, preservation, and/or restoration of wetlands and to judge the effectiveness of these regulatory efforts in preserving associated ecosystem services, wetlands must be routinely monitored. Wetland hydrology is the most important abiotic factor controlling wetland function and extent, and should therefore be a vital part of any wetland mapping or monitoring program. This information is critical to informing the wetland regulation debate (e.g., the impact of the “significant nexus” concept on the jurisdiction of the Clean Water Act) ongoing within the federal government. Broad-scale forested wetland hydrology has been difficult to study with conventional remote sensing methods and in situ data collection is resource prohibitive. Airborne Light Detection and Ranging (LiDAR) is a new and rapidly developing technology. LiDAR data have mainly been used to derive information on elevation. However, the intensity (amplitude) of the signal used to derive information on elevation has the potential to significantly improve the ability to remotely monitor forested wetland hydrology. Few studies have taken advantage of LiDAR intensity for image classification and none have demonstrated the ability of LiDAR intensity data to map and monitor hydrology in forested wetlands. A comparison between LiDAR intensity data collected during peak hydrologic expression in a series of forested wetlands on the Eastern Shore of Maryland and detailed in situ data demonstrate the strong potential of LiDAR intensity data for this application (>96% overall accuracy). The relative ability of LiDAR intensity data for forest hydrology mapping was compared with that of a false color near-infrared aerial photograph collected coincident with the LiDAR intensity (70% overall accuracy) – currently the most commonly used method for wetland mapping and a wetness index derived from a digital elevation model. The potential of LiDAR intensity data is strong for addressing the regulatory status of wetlands and measuring the delivery of ecosystem services.