Mapping Wetland Vegetation with LiDAR in Everglades National Park, Florida, USA PDF Download
Are you looking for read ebook online? Search for your book and save it on your Kindle device, PC, phones or tablets. Download Mapping Wetland Vegetation with LiDAR in Everglades National Park, Florida, USA PDF full book. Access full book title Mapping Wetland Vegetation with LiDAR in Everglades National Park, Florida, USA by Georgia H. De Stoppelaire. Download full books in PDF and EPUB format.
Author: Georgia H. De Stoppelaire Publisher: ISBN: Category : Coastal zone management Languages : en Pages : 153
Book Description
Knowledge of the geospatial distribution of vegetation is fundamental for resource management. The objective of this study is to investigate the possible use of airborne LIDAR (light detection and ranging) data to improve classification accuracy of high spatial resolution optical imagery and compare the ability of two classification algorithms to accurately identify and map wetland vegetation communities. In this study, high resolution imagery integrated with LIDAR data was compared jointly and alone; and the nearest neighbor (NN) and machine learning random forest (RF) classifiers were assessed in semi-automated geographic object-based image analysis (GEOBIA) approaches for classification accuracy of heterogeneous vegetation assemblages at Everglades National Park, FL, USA. Within the 145 ha study area, five dominant vegetation communities that establish the vegetation pattern were mapped: Subtropical hardwood forest, Slash pine with hardwoods, Pine savanna, Hardwood scrub, and Sawgrass. A LIDAR-derived canopy height model (CHM) was produced at 1 m spatial resolution and integrated with Digital Orthophoto Quarter-Quadrangle (DOQQ) optical imagery. Object-based segmentation was performed using the multi-resolution segmentation algorithm at optimal scale based on a global score. In a series of 42 experiments using the NN and RF classifiers, two data schemes were tested: fused data and optical imagery. Inclusion of additional first-order statistical features were also tested in the RF experiments. Results showed that the fused data produced significantly higher classification accuracy at the 95% confidence level than optical imagery alone in both sets of experiments. Among the classifiers, data schemes, and feature sets tested, the NN experiment using fused data with features of mean spectral values and mean CHM elevation values produced the highest overall accuracy (OA) of 83.6% and Kappa of 0.782, while the highest accuracy RF experiment produced an OA of 75.73% and Kappa of 0.695. Pairwise comparison of error matrices for the highest accuracy NN and RF experimental results were significantly different at the 95% confidence level with a Z score of 3.26. Findings show that the integration of LIDAR significantly improved classification accuracy of high spatial resolution optical imagery to identify and map wetland vegetation. Results from this study demonstrate fused LIDAR and high resolution imagery can be used to accurately map wetland vegetation assemblages in a repeatable, semi-automated GEOBIA approach.
Author: Georgia H. De Stoppelaire Publisher: ISBN: Category : Coastal zone management Languages : en Pages : 153
Book Description
Knowledge of the geospatial distribution of vegetation is fundamental for resource management. The objective of this study is to investigate the possible use of airborne LIDAR (light detection and ranging) data to improve classification accuracy of high spatial resolution optical imagery and compare the ability of two classification algorithms to accurately identify and map wetland vegetation communities. In this study, high resolution imagery integrated with LIDAR data was compared jointly and alone; and the nearest neighbor (NN) and machine learning random forest (RF) classifiers were assessed in semi-automated geographic object-based image analysis (GEOBIA) approaches for classification accuracy of heterogeneous vegetation assemblages at Everglades National Park, FL, USA. Within the 145 ha study area, five dominant vegetation communities that establish the vegetation pattern were mapped: Subtropical hardwood forest, Slash pine with hardwoods, Pine savanna, Hardwood scrub, and Sawgrass. A LIDAR-derived canopy height model (CHM) was produced at 1 m spatial resolution and integrated with Digital Orthophoto Quarter-Quadrangle (DOQQ) optical imagery. Object-based segmentation was performed using the multi-resolution segmentation algorithm at optimal scale based on a global score. In a series of 42 experiments using the NN and RF classifiers, two data schemes were tested: fused data and optical imagery. Inclusion of additional first-order statistical features were also tested in the RF experiments. Results showed that the fused data produced significantly higher classification accuracy at the 95% confidence level than optical imagery alone in both sets of experiments. Among the classifiers, data schemes, and feature sets tested, the NN experiment using fused data with features of mean spectral values and mean CHM elevation values produced the highest overall accuracy (OA) of 83.6% and Kappa of 0.782, while the highest accuracy RF experiment produced an OA of 75.73% and Kappa of 0.695. Pairwise comparison of error matrices for the highest accuracy NN and RF experimental results were significantly different at the 95% confidence level with a Z score of 3.26. Findings show that the integration of LIDAR significantly improved classification accuracy of high spatial resolution optical imagery to identify and map wetland vegetation. Results from this study demonstrate fused LIDAR and high resolution imagery can be used to accurately map wetland vegetation assemblages in a repeatable, semi-automated GEOBIA approach.
Author: Caiyun Zhang Publisher: CRC Press ISBN: 0429530781 Category : Science Languages : en Pages : 409
Book Description
This book explores the applicability of multiple remote sensors to acquire information relevant to restoration and conservation efforts in wetlands using data collected from airborne and space multispectral/hyperspectral sensors, light detection and ranging (LiDAR), Unmanned Aircraft Systems (UAS), and a hand-held spectroradiometer. This book also examines digital data processing techniques such as object-based image analysis, machine learning, texture analysis, and data fusion. After an introduction to the Everglades and to remote sensing, the book is divided into four parts based on the sensor systems used. There are chapters on vegetation mapping, biomass and water quality modeling, applications of hyperspectral data for plant stress analysis and coral reef mapping, studies of airborne LiDAR data for coastal vulnerability analysis and DEM improvement, as well as chapters that explore a fusion of multiple sensors for different datasets. Features Introduces concepts, theories, and advanced processing techniques A complete introduction of machine learning, object-based image analysis, data fusion, and ensemble analysis techniques in processing data from multiple remote sensors Explains how multiple remote sensing systems are applied in the wetland ecosystems of Florida The author had been teaching and using both systems and her research is widely recognized Multi-sensor System Applications in the Everglades Ecosystems provides a comprehensive application of remote sensing techniques in the Florida Everglades and its coastal ecosystems. It will prove an invaluable resource for the restoration and conservation of the Florida Everglades and beyond, for global wetlands in general. Any professional, scientist, engineer, or student working with remote sensing and wetland ecosystems will reap enormous benefits from this book.
Author: Ravi Shankar Dwivedi Publisher: CRC Press ISBN: 1040121829 Category : Technology & Engineering Languages : en Pages : 777
Book Description
Timely and reliable information on natural resources, regarding their potential and limitations, is a prerequisite for sustainable development. Geospatial technologies offer immense potential in providing such information in a timely and cost-effective manner. Using orbital sensors data in conjunction with airborne and proximal sensors data to generate information on soils and agricultural resources, forests, mineral resources, fossil fuel, wetlands, water resources, and marine resources, this book focuses on the advancements in technologies applicable to managing these resources. It addresses global issues like climate change and land degradation neutrality and introduces spatial data infrastructure (SDI) as a mechanism for sharing geospatial data. This book also provides an in-depth discussion on drones, crowdsourcing, cloud computing, Internet of Things, machine learning, and their applications. FEATURES Contains a comprehensive resource on the latest developments in geospatial technologies and their use in monitoring natural resources, productivity mapping, and modeling Explains the geo-computation methods and online algorithm developments Includes clear guidance on how best to use geospatial data for various applications Discusses case studies from a variety of fields and current trends in the management of natural resources Provides future scenarios concerning platforms, sensors, data analysis, and interpretation techniques This book is written for remote sensing and GIS professionals in environmental institutions and government who are involved in natural resource management projects. Senior undergraduate and graduate-level students in Earth sciences, geography, or environmental management can also use this text for supplementary reading.
Author: Zoltan Szantoi Publisher: ISBN: Category : Languages : en Pages :
Book Description
ABSTRACT: This study developed a classification and discrimination methodology with quantifiable accuracies for wetland mapping, focusing on at risk plant communities in marl prairie and marsh areas of the Everglades National Park, Florida. Accurately and rapidly mapping the baseline spatial distribution of these communities with classification algorithms using accessible and temporally frequent imagery is critical to better assess changes in wet graminoid/sedge communities and their effects on the ecotone, wildlife habitats and overall ecological change. Maximum likelihood (ML), Decision Tree (DT) and Artificial Neural Network (ANN) classifiers were tested on high resolution aerial imagery (30.5 cm) of the Everglades National Park, Florida.
Author: National Research Council Publisher: National Academies Press ISBN: 0309103355 Category : Political Science Languages : en Pages : 250
Book Description
This report is the first in a congressionally mandated series of biennial evaluations of the progress being made by the Comprehensive Everglades Restoration Plan (CERP), a multibillion-dollar effort to restore historical water flows to the Everglades and return the ecosystem closer to its natural state, before it was transformed by drainage and by urban and agricultural development. The Restoration plan, which was launched in 1999 by the U.S. Army Corps of Engineers and the South Florida Water Management District, includes more than 40 major projects that are expected to be completed over the next three decades. The report finds that progress has been made in developing the scientific basis and management structures needed to support a massive effort to restore the Florida Everglades ecosystem. However, some important projects have been delayed due to several factors including budgetary restrictions and a project planning process that that can be stalled by unresolved scientific uncertainties. The report outlines an alternative approach that can help the initiative move forward even as it resolves remaining scientific uncertainties. The report calls for a boost in the rate of federal spending if the restoration of Everglades National Park and other projects are to be completed on schedule.
Author: Fred H. Sklar Publisher: Springer Science & Business Media ISBN: 9400900015 Category : Science Languages : en Pages : 539
Book Description
PREFACE Within the Florida Everglades, tree islands, which cover only a small percentofthis ecosystem, historically have provided essential habitat for a wide variety ofterrestrial and amphibious plants, birds, and animals. These tree islands, however, have been one ofits least studied features. Because of their less flood tolerant vegetation, tree islands are one ofthe most sensitive components ofthe Everglades to changes in hydrology, and many tree islands have been lost during periods when water levels have been abnormally high or low. Their sensitivity to water level changes makes tree islands potentially one ofthe best and surest measures ofthe overall hydrologic health of the Everglades. Consequently, the maintenance of healthy, functioning tree islands and the restoration ofthose that have been lost will be an important performance measures that will be used tojudge the success ofthe Comprehensive Everglades Restoration Plan (CERP). A symposium, Tree Islands ofthe Everglades, was held on July 14 and 15, 1998 at Florida Atlantic University, Boca Raton, Florida. It was sponsored by Florida Center for Environmental Studies and the South Florida WaterManagement District. This was the first scientific meeting ever devoted to tree islands. The organizers of this symposium were Drs. Arnold van der Valk, Florida Center for Environmental Studies and Iowa State University, Fred Sklar, South Florida Water Management District, and Wiley Kitchens, United States Geological Survey.
Author: National Aeronautics and Space Administration (NASA) Publisher: Createspace Independent Publishing Platform ISBN: 9781724281272 Category : Languages : en Pages : 28
Book Description
The success of remotely mapping wetland vegetation of the southwestern coast of Florida is examined. A computerized technique to process aircraft and LANDSAT multispectral scanner data into vegetation classification maps was used. The cost effectiveness of this mapping technique was evaluated in terms of user requirements, accuracy, and cost. Results indicate that mangrove communities are classified most cost effectively by the LANDSAT technique, with an accuracy of approximately 87 percent and with a cost of approximately 3 cent per hectare compared to $46.50 per hectare for conventional ground survey methods. Butera, M. K. Stennis Space Center NASA-TP-1553, S-492 RTOP 177-52-83-13
Author: Georgia H. De Stoppelaire
Author: Georgia H. De Stoppelaire
Author: Caiyun Zhang
Author: Ravi Shankar Dwivedi
Author: 
Author: Zoltan Szantoi
Author: National Research Council
Author: John C. Brock
Author: Fred H. Sklar
Author: National Aeronautics and Space Administration (NASA)
Author: Raymond W. Schaffranek