scholarly journals Classification of Typha-dominated wetlands using airborne hyperspectral imagery along Lake Ontario, USA

2021 ◽  
Vol 24 (2) ◽  
pp. 140-155
Author(s):  
Glenn M. Suir ◽  
Douglas A. Wilcox ◽  
Molly Reif
Keyword(s):  
Adaptive Management ◽  
Coastal Wetland ◽  
Hyperspectral Imagery ◽  
Monitoring Program ◽  
Lake Ontario ◽  
Training Data ◽  
Army Corps ◽  
Wetland Classification ◽  
Spatial And Temporal Scales ◽  
Sensing Applications

Abstract Shoreline wetlands along Lake Ontario are valuable, multi-functional resources that have historically provided large numbers of important ecosystem goods and services. However, alterations to the lake’s natural hydrologic regime have impacted traditional meadow marsh in the wetlands, resulting in competition and colonization by dense and aggressive Typha angustifolia and Typha x glauca (Cattails). The shift to a Typha-dominated landscape resulted in an array of negative impacts, including increased Typha density, substantial decreases in plant species richness and diversity, and altered habitat and changes in associated ecosystem services. Successful long-term adaptive management of these wetland resources requires timely and accurate monitoring. Historically, wetland landscapes have been surveyed and mapped using field-based surveys and/or photointerpretation. However, given their resource- and cost-intensive nature, these methods are often prohibitively time- and labor-consuming or geographically limited. Other remote sensing applications can provide more rapid and efficient assessments when evaluating wetland change trajectories or analyzing direct and indirect impacts across larger spatial and temporal scales. The primary goal of this study was to develop and describe methodology using U.S. Army Corps of Engineers National Coastal Mapping Program hyperspectral imagery, light detection and ranging data, and high-spatial resolution true-color imagery to provide updated wetland classifications for Lake Ontario coastal wetlands. This study used existing field-collected vegetation survey data (Great Lakes Coastal Wetland Monitoring Program), ancillary imagery, and existing classification information as training data for a supervised classification approach. These data were used along with a generalized wetland schema (classes based on physical and biological gradients: elevation, Typha, meadow marsh, mixed emergent, upland vegetation) to generate wetland classification data with Kappa values near 0.85. Ultimately, these data and methods provide helpful knowledge elements that will allow for more efficient inventorying and monitoring of Great Lake resources, forecasting of resource condition and stability, and adaptive management strategies.

scholarly journals Swin Transformer for Complex Coastal Wetland Classification Using the Integration of Sentinel-1 and Sentinel-2 Imagery

Water ◽  
2022 ◽  
Vol 14 (2) ◽  
pp. 178
Author(s):  
Ali Jamali ◽  
Masoud Mahdianpari
Keyword(s):  
Remote Sensing ◽  
Language Processing ◽  
Coastal Wetland ◽  
Remote Sensing Data ◽  
Machine Learning Algorithms ◽  
Coastal Marsh ◽  
Forested Wetland ◽  
Wetland Classification ◽  
Sensing Applications

The emergence of deep learning techniques has revolutionized the use of machine learning algorithms to classify complicated environments, notably in remote sensing. Convolutional Neural Networks (CNNs) have shown considerable promise in classifying challenging high-dimensional remote sensing data, particularly in the classification of wetlands. State-of-the-art Natural Language Processing (NLP) algorithms, on the other hand, are transformers. Despite the fact that transformers have been utilized for a few remote sensing applications, they have not been compared to other well-known CNN networks in complex wetland classification. As such, for the classification of complex coastal wetlands in the study area of Saint John city, located in New Brunswick, Canada, we modified and employed the Swin Transformer algorithm. Moreover, the developed transformer classifier results were compared with two well-known deep CNNs of AlexNet and VGG-16. In terms of average accuracy, the proposed Swin Transformer algorithm outperformed the AlexNet and VGG-16 techniques by 14.3% and 44.28%, respectively. The proposed Swin Transformer classifier obtained F-1 scores of 0.65, 0.71, 0.73, 0.78, 0.82, 0.84, and 0.84 for the recognition of coastal marsh, shrub, bog, fen, aquatic bed, forested wetland, and freshwater marsh, respectively. The results achieved in this study suggest the high capability of transformers over very deep CNN networks for the classification of complex landscapes in remote sensing.

scholarly journals Habitat Restoration Across Large Areas: Assessing Wildlife Responses in the Clearwater Basin, Idaho

2004 ◽  
Vol 19 (2) ◽  
pp. 123-132 ◽  
Author(s):  
L. K. Svancara ◽  
G. Servheen ◽  
W. Melquist ◽  
D. Davis ◽  
J. M. Scott
Keyword(s):  
Adaptive Management ◽  
Habitat Management ◽  
Monitoring Program ◽  
Population Data ◽  
Management Policy ◽  
Management Approach ◽  
Past Century ◽  
Spatial And Temporal Scales ◽  
Wildlife Species ◽  
Management Actions

Abstract Over the past century, fire suppression and prevention have altered disturbance regimes across the Pacific Northwest, resulting in a significant divergence of historical and current conditions in forested habitats. To address this continuing trend in habitat changes and begin restoring historical patterns of disturbance, the Clearwater Basin Elk Habitat Initiative (CEI) proposes relatively extensive management actions in the Clearwater basin of north-central Idaho. We attempted to evaluate potential effects of such management actions on selected wildlife species using extant data sets and suggest ways to improve such projects with respect to a multispecies and adaptive management approach. Although there is increased interest in ecosystem management over large areas, the increased scale of analysis and implementation require a substantial increase in the level of species information beyond what currently exists. We conclude that baseline information required for an effective multispecies land-management policy in the Clearwater basin does not exist for many terrestrial wildlife species. To implement a true multispecies or ecosystem approach, wildlife and land managers should cooperate to increase existing population data and modeling efforts for wildlife species in the basin and develop a sustainable monitoring program to evaluate habitat management changes and their influence on wildlife populations within the context of adaptive management theory. Management actions to restore disturbance patterns should attempt spatial and temporal scales that are biologically relevant to the population ecology of species being affected. West. J. Appl. For. 19(2): 123–132.

scholarly journals Coastal Wetland Classification with Deep U-Net Convolutional Networks and Sentinel-2 Imagery: A Case Study at the Tien Yen Estuary of Vietnam

2020 ◽  
Vol 12 (19) ◽  
pp. 3270
Author(s):  
Kinh Bac Dang ◽  
Manh Ha Nguyen ◽  
Duc Anh Nguyen ◽  
Thi Thanh Hai Phan ◽  
Tuan Linh Giang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Coastal Wetland ◽  
Coastal Hazards ◽  
Support Vector ◽  
Wetland Classification ◽  
Natural Wetland ◽  
Convolutional Networks ◽  
Transition Areas ◽  
Wetland Types ◽  
Sentinel 2

The natural wetland areas in Vietnam, which are transition areas from inland and ocean, play a crucial role in minimizing coastal hazards; however, during the last two decades, about 64% of these areas have been converted from the natural wetland to the human-made wetland. It is anticipated that the conversion rate continues to increase due to economic development and urbanization. Therefore, monitoring and assessment of the wetland are essential for the coastal vulnerability assessment and geo-ecosystem management. The aim of this study is to propose and verify a new deep learning approach to interpret 9 of 19 coastal wetland types classified in the RAMSAR and MONRE systems for the Tien Yen estuary of Vietnam. Herein, a Resnet framework was integrated into the U-Net to optimize the performance of the proposed deep learning model. The Sentinel-2, ALOS-DEM, and NOAA-DEM satellite images were used as the input data, whereas the output is the predefined nine wetland types. As a result, two ResU-Net models using Adam and RMSprop optimizer functions show the accuracy higher than 85%, especially in forested intertidal wetlands, aquaculture ponds, and farm ponds. The better performance of these models was proved, compared to Random Forest and Support Vector Machine methods. After optimizing the ResU-Net models, they were also used to map the coastal wetland areas correctly in the northeastern part of Vietnam. The final model can potentially update new wetland types in the southern parts and islands in Vietnam towards wetland change monitoring in real time.

scholarly journals Mapping Coastal Wetland Biomass from High Resolution Unmanned Aerial Vehicle (UAV) Imagery

2019 ◽  
Vol 11 (5) ◽  
pp. 540 ◽  
Author(s):  
Cheryl Doughty ◽  
Kyle Cavanaugh
Keyword(s):  
Salt Marsh ◽  
High Resolution ◽  
Aboveground Biomass ◽  
Vegetation Index ◽  
Coastal Wetland ◽  
Spatial Scales ◽  
Biomass Estimation ◽  
Spatial And Temporal Variability ◽  
Estimation Model ◽  
Spatial And Temporal Scales

Salt marsh productivity is an important control of resiliency to sea level rise. However, our understanding of how marsh biomass and productivity vary across fine spatial and temporal scales is limited. Remote sensing provides a means for characterizing spatial and temporal variability in marsh aboveground biomass, but most satellite and airborne sensors have limited spatial and/or temporal resolution. Imagery from unmanned aerial vehicles (UAVs) can be used to address this data gap. We combined seasonal field surveys and multispectral UAV imagery collected using a DJI Matrice 100 and Micasense Rededge sensor from the Carpinteria Salt Marsh Reserve in California, USA to develop a method for high-resolution mapping of aboveground saltmarsh biomass. UAV imagery was used to test a suite of vegetation indices in their ability to predict aboveground biomass (AGB). The normalized difference vegetation index (NDVI) provided the strongest correlation to aboveground biomass for each season and when seasonal data were pooled, though seasonal models (e.g., spring, r2 = 0.67; RMSE = 344 g m−2) were more robust than the annual model (r2 = 0.36; RMSE = 496 g m−2). The NDVI aboveground biomass estimation model (AGB = 2428.2 × NDVI + 120.1) was then used to create maps of biomass for each season. Total site-wide aboveground biomass ranged from 147 Mg to 205 Mg and was highest in the spring, with an average of 1222.9 g m−2. Analysis of spatial patterns in AGB demonstrated that AGB was highest in intermediate elevations that ranged from 1.6–1.8 m NAVD88. This UAV-based approach can be used aid the investigation of biomass dynamics in wetlands across a range of spatial scales.

Sign in / Sign up

Export Citation Format

Share Document