Wetland Inventory Research Articles

Mapping Coastal Wetland Changes from 1985 to 2022 in the US Atlantic and Gulf Coasts using Landsat Time Series and National Wetland Inventories

Mapping Coastal Wetland Changes from 1985 to 2022 in the US Atlantic and Gulf Coasts using Landsat Time Series and National Wetland Inventories

A “Region-Specific Model Adaptation (RSMA)”-Based Training Data Method for Large-Scale Land Cover Mapping

An accurate and historical land cover monitoring dataset for Alaska could provide fundamental information for a range of studies, such as conservation habitats, biogeochemical cycles, and climate systems, in this distinctive region. This research addresses challenges associated with the extraction of training data for timely and accurate land cover classifications in Alaska over longer time periods (e.g., greater than 10 years). Specifically, we designed the “Region-Specific Model Adaptation (RSMA)” method for training data. The method integrates land cover information from the National Land Cover Database (NLCD), LANDFIRE’s Existing Vegetation Type (EVT), and the National Wetlands Inventory (NWI) and machine learning techniques to generate robust training samples based on the Anderson Level II classification legend. The assumption of the method is that spectral signatures vary across regions because of diverse land surface compositions; however, despite these variations, there are consistent, collective land cover characteristics that span the entire region. Building upon this assumption, this research utilized the classification power of deep learning algorithms and the generalization ability of RSMA to construct a model for the RSMA method. Additionally, we interpreted existing vegetation plot information for land cover labels as validation data to reduce inconsistency in the human interpretation. Our validation results indicate that the RSMA method improved the quality of the training data derived solely from the NLCD by approximately 30% for the overall accuracy. The validation assessment also demonstrates that the RSMA method can generate reliable training data on large scales in regions that lack sufficient reliable data.

The feasibility of using national‐scale datasets for classifying wetlands in Arizona with machine learning

AbstractThe advent of machine learning techniques has led to a proliferation of landscape classification products. These approaches can fill gaps in wetland inventories across the United States (U.S.) provided that large reference datasets are available to develop accurate models. In this study, we tested the feasibility of expediting the classification process by sourcing requisite training and testing data from existing national‐scale land cover maps instead of customized sample sets. We created a single map of water and wetland presence by intersecting water and wetland classes from available land cover products (National Wetland Inventory, Gap Analysis Project, National Land Cover Database and Dynamic Surface Water Extent) across the U.S. state of Arizona, which has fewer wetland‐specific mapping products than other parts of the U.S. We derived classified samples for four wetland classes from the combined map: open water, herbaceous wetlands, wooded wetlands and non‐wetland cover. In Google Earth Engine, we developed a random forest model that combined the training data with spatial predictor variables, including vegetation greenness indices, wetness indices, seasonal index variation, topographic parameters and vegetation height metrics. Results show that the final model separates the four classes with an overall accuracy of 86.2%. The accuracy suggests that existing datasets can be effectively used to compile machine learning training samples to map wetlands in arid landscapes in the U.S. These methods hold promise for the generation of wetland inventories at more frequent intervals, which could allow more nuanced investigations of wetland change over time in response to anthropogenic and climatic drivers.

Groundwater driven carbon fluxes in a restored coastal saltmarsh wetland: Implications for coastal wetland restoration

Coastal wetlands play a crucial role in the global carbon cycle, yet they have been extensively degraded over the past decade. Though restoration efforts are underway worldwide, there is limited understanding of the role groundwater plays in transporting dissolved carbon within restored wetlands. Here, we address this knowledge gap by investigating water and carbon fluxes in the restored Tomago Wetlands in coastal NSW, Australia. We aim to 1) quantify surface water exports of the main carbon components, 2) estimate greenhouse gas emissions from the aquatic parts of the wetland, and 3) determine the contribution of groundwater discharge to the surface water carbon export and greenhouse gas emissions from the wetland. Sampling took place following a wet period and a freshening event. A radon mass balance model estimated groundwater discharge contributing 10 % to the surface water flow in the restored wetland. The wetland was a source of four of the five main carbon components including DOC, DIC, alkalinity, and CO2, with most of the exported carbon being in the form of alkalinity. Surface water DOC, DIC, alkalinity, and CO2 exports were, 1.3 ± 0.5, 35.5 ± 13.1, 39.4 ± 14.6 and 6.5 ± 2.1 mmol/m2/d, respectively. The average water to atmosphere flux of CO2 and CH4 were 104 ± 209 (mmol/m2/d) and 21 ± 42 (µmol/m2/d). Groundwater-driven carbon fluxes contributed 100 % of the surface water DOC, 30 % of the DIC, 15 % of the alkalinity exports, and 5 % of the overall CO2 losses (lateral aqueous export + atmospheric emissions) from the wetland, with minor contributions to CH4. Carbon exports from both the wetland and groundwater observed in this restored wetland were found to be lower than those reported in the literature for natural or mature wetlands. This would have important implications when developing carbon accounting methods. Overall, our findings highlight the importance of groundwater in carbon transport within restored wetlands and emphasise the need for inclusion of shallow groundwater dynamics in carbon budgets and inventories of coastal wetlands that have or will undergo restoration.

Advancements in mapping areas suitable for wetland habitats across the conterminous United States

Wetland habitats provide critical ecosystem services to the surrounding landscape, including nutrient and pollutant retention, flood mitigation, and carbon storage. Wetland connectivity to water bodies and related ecosystems is critical in habitat sustainability, but there are limited resources for landscape-level wetland planning. Considering the network connectivity of an ecosystem type can derive different benefits to the natural and built environment, as well as human health. The value that wetlands provide, along with incentive programs and conservation goals mandated by the government require new and improved wetland spatial data. Utilizing high quality, publicly available data, this study finds that the amount of land in the United States that could support built or restored wetlands is more than double the area of mapped existing wetlands.This study uses 17 input variables (i.e., features extracted from remotely sensed data and auxiliary datasets) at the 10-m resolution and the National Wetlands Inventory to train a random forest model to identify areas that may support a wetland habitat, or potential wetland areas. Models were calculated for each of 18 two-digit hydrologic units that encompass the conterminous United States, and model overall accuracy ranged from 78.0 % to 89.8 %. The models predicted that 21.1 % of the conterminous United States can be categorized as potential wetland area.Selecting input variables to predict areas with wetland potential, rather than to identify existing wetlands, using the random forest algorithm can be transferred to other locations, scales, and ecosystem types. Visualizing potential wetland areas using input data at the 10-m resolution and enhanced methodology improves previous work, as even slight changes in topography, soils, and landscape features can determine ecosystem connections. This product can be used to better place wetland restoration projects to serve ecosystem- and community-wide health by ensuring ecosystem success and targeting areas that face increased climate change impacts.

Hotspots of wetland loss to impervious surfaces in the conterminous United States

In this study, a first wall-to-wall comparison between the National Wetlands Inventory (NWI) and the National Land Cover Database (NLCD) was conducted across the entire conterminous United States (CONUS) to evaluate U.S. wetland loss conditions. Annually, around 26 km2 of wetlands are lost to impervious surfaces across the CONUS. Spatially, wetland loss is not evenly distributed, with 90 % of losses occurring in only 9 % of the land area, forming hotspots around expanding urban regions such as Houston, Jacksonville, and Naples. Over the past few decades, Florida experienced the highest wetland loss (5.73 km2/year) among all states, while Houston had the most wetland loss (2.54 km2/year) among all metropolitan regions. Stepwise multiple regression models identified population growth and its associated demand for new housing as the major drivers for wetland loss. Wetland loss per population increase is the highest (>15 m2/person) in most metropolitan regions around the East Coast and Gulf of Mexico. Unfortunately, current wetland loss hotspots will likely suffer further losses in future decades due to projected population growth, with Houston, Cape Coral, and Miami metropolitan regions having the greatest projected wetland loss of 89.15 km2, 34.35 km2, and 28.20 km2, respectively. This study has identified wetland loss hotspots and their drivers across the U.S. that were not possible in previous sample-based studies. The findings are critical in wetland management and protection across the U.S.

Comparison of coastal wetland inventories for representative sites in the United States and implications for change detection

Comparison of coastal wetland inventories for representative sites in the United States and implications for change detection

A hydrologic and land cover‐based habitat model for use in bog turtle (Glyptemys muhlenbergii) conservation

Abstract The endangered bog turtle (Glyptemys muhlenbergii) exemplifies issues related to rare species conservation; presence surveys have low detection and variables used in habitat models can lack relevance to established biological relationships of the species to its environment. The species' use of groundwater saturated soils and stream networks as core habitat and dispersal corridors has been documented. Less is known about the landscape factors that promote the formation and persistence of wetlands used for core habitat. A GIS‐based resource selection function was developed to predict bog turtle habitat use. Trained on 1 ha plots centred on occupied sites and pseudo‐absent plots constrained to areas within 56.4 m to stream network centrelines, the model tested the capacity of a topographic wetness index (TWI), stream order and soil, wetland and land cover type to predict the presence of suitable habitat and turtle occupancy. Landscape variables were sampled at 10 m resolution, but variable selection and model performance were analysed at 100 m resolution to maintain a biologically relevant 1 ha habitat scale and accommodate the resolution of other variables. Suitable habitat and turtle presence were best predicted by intermediate to high values of TWI, land cover with low vegetation height and wetlands, second‐ and third‐order streams and the occurrence of mapped National Wetland Inventory polygons and hydric soils. Very high values of TWI were negatively associated with habitat suitability. The area under the curve of the best model was 0.833. Suitable habitat was found on 88% of 55 independent sites selected using the model and nine new occupied sites were confirmed. Model error is discussed with consideration of human‐altered drainage networks on the agricultural landscape.

Plant distribution and conservation in mediterranean islands’ lentic wetlands: there is more than aquatic species

This study investigates the distribution and conservation status of plant species in lentic wetlands of Mediterranean islands. Field data on vascular macrophytes were collected from the online national databases of the “Conservation of the island wetlands of the Mediterranean Basin” (MedIsWet). A comprehensive inventory of wetlands was conducted, and information on the main anthropogenic impacts and conservation priorities was compiled. In addition to all vascular hygro- and hydrophytes, we retained information on those species typically adapted to environments with intermittent or temporary wetness. A significant knowledge gap regarding the distribution and conservation of plant species in Mediterranean wetlands was found. A lower endemism rate was observed particularly in coastal wetlands. However, approximately 25% of species, mainly with relatively wide distribution, were identified as endangered or with unknown conservation status. Including all macrophytes adapted to seasonal wetness is crucial when considering wetland conservation efforts. The research emphasised the importance of considering diverse wetland types, including seasonal and permanent, natural and artificial, for effective plant conservation. Artificial wetlands emerged as potential habitats with considerable biodiversity conservation value. This study provides a comprehensive inventory of wetlands and valuable insights into the distribution, ecology, and conservation relevance of aquatic macrophytes in Mediterranean islands. The research enhances our understanding of biogeographic patterns and processes, offers critical information for the management and conservation of Mediterranean island wetlands, and presents a replicable approach that can be applied to other wetland contexts.

Simulation of the Impacts of Sea-Level Rise on Coastal Ecosystems in Benin Using a Combined Approach of Machine Learning and the Sea Level Affecting Marshes Model

Sea-level rise in Benin coastal zones leads to risks of erosion and flooding, which have significant consequences on the socio-economic life of the local population. In this paper, erosion, flood risk, and greenhouse gas sequestration resulting from sea-level rise in the coastal zone of the Benin coast were assessed with the Sea Level Affecting Marshes Model (SLAMM) using ArcGIS Pro 3.1 tools. The input features used were the Digital Elevation Map (DEM), the National Wetland Inventory (NWI) categories, and the slope of each cell. National Wetland Inventory (NWI) categories were then created using Support Vector Machines (SVMs), a supervised machine learning technique. The research simulated the effects of a 1.468 m sea-level rise in the study area from 2021 to 2090, considering wetland types, marsh accretion, wave erosion, and surface elevation changes. The largest land cover increases were observed in Estuarine Open Water and Open Ocean, expanding by approximately 106.2 hectares across different sea-level rise scenarios (RCP 8.5_Upper Limit). These gains were counterbalanced by losses of approximately 106.2 hectares in Inland Open Water, Ocean Beaches, Mangroves, Regularly Flooded Marsh, Swamp, Undeveloped, and Developed Dryland. Notably, Estuarine Open Water (97.7 hectares) and Open Ocean (8.5 hectares) experienced the most significant expansion, indicating submergence and saltwater intrusion by 2090 due to sea-level rise. The largest reductions occurred in less tidally influenced categories like Inland Open Water (−81.4 hectares), Ocean Beach (−7.9 hectares), Swamp (−5.1 hectares), Regularly Flooded Marsh (−4.6 hectares), and Undeveloped Dryland (−2.9 hectares). As the sea-level rises by 1.468 m, these categories are expected to be notably diminished, with Estuarine Open Water and Open Ocean becoming dominant. Erosion and flooding in the coastal zone are projected to have severe adverse impacts, including a gradual decline in greenhouse gas sequestration capacity. The outputs of this research will aid coastal management organizations in evaluating the consequences of sea-level rise and identifying areas with high mitigation requirements.

The Wetland Intrinsic Potential tool: mapping wetland intrinsic potential through machine learning of multi-scale remote sensing proxies of wetland indicators

Abstract. Accurate, unbiased wetland inventories are critical to monitor and protect wetlands from future harm or land conversion. However, most wetland inventories are constructed through manual image interpretation or automated classification of multi-band imagery and are biased towards wetlands that are easy to directly detect in aerial and satellite imagery. Wetlands that are obscured by forest canopy, that occur ephemerally, and that have no visible standing water are, therefore, often missing from wetland maps. To aid in the detection of these cryptic wetlands, we developed the Wetland Intrinsic Potential (WIP) tool, based on a wetland-indicator framework commonly used on the ground to detect wetlands through the presence of hydrophytic vegetation, hydrology, and hydric soils. Our tool uses a random forest model with spatially explicit input variables that represent all three wetland indicators, including novel multi-scale topographic indicators that represent the processes that drive wetland formation, to derive a map of wetland probability. With the ability to include multi-scale topographic indicators that help identify cryptic wetlands, the WIP tool can identify areas conducive to wetland formation while providing a flexible approach that can be adapted to diverse landscapes. For a study area in the Hoh River watershed in western Washington, USA, classification of the output probability with a threshold of 0.5 provided an overall accuracy of 91.97 %. Compared to the National Wetlands Inventory, the classified WIP tool output identified over 2 times the wetland area and reduced errors of omission from 47.5 % to 14.1 % but increased errors of commission from 1.9 % to 10.5 %. The WIP tool is implemented as an ArcGIS toolbox using a combination of R and Python scripts.

Rapid Large-Scale Wetland Inventory Update Using Multi-Source Remote Sensing

Rapid impacts from both natural and anthropogenic sources on wetland ecosystems underscore the need for updating wetland inventories. Extensive up-to-date field samples are required for calibrating methods (e.g., machine learning) and validating results (e.g., maps). The purpose of this study is to design a dataset generation approach that extracts training data from already existing wetland maps in an unsupervised manner. The proposed method utilizes the LandTrendr algorithm to identify areas least likely to have changed over a seven-year period from 2016 to 2022 in Minnesota, USA. Sentinel-2 and Sentinel-1 data were used through Google Earth Engine (GEE), and sub-pixel water fraction (SWF) and normalized difference vegetation index (NDVI) were considered as wetland indicators. A simple thresholding approach was applied to the magnitude of change maps to identify pixels with the most negligible change. These samples were then employed to train a random forest (RF) classifier in an object-based image analysis framework. The proposed method achieved an overall accuracy of 89% with F1 scores of 91%, 81%, 88%, and 72% for water, emergent, forested, and scrub-shrub wetland classes, respectively. The proposed method offers an accurate and cost-efficient method for updating wetland inventories as well as studying areas impacted by floods on state or even national scales. This will assist practitioners and stakeholders in maintaining an updated wetland map with fewer requirements for extensive field campaigns.

Stakeholders’ composition, knowledge and use of earth observation data in wetland ecological assessments, Zimbabwe

The study analyses stakeholders’ knowledge and use of earth observation (EO) derived data in the assessment and monitoring of Driefontein and Intunjambili wetlands’ ecological conditions. Data was collected using a questionnaire survey administered to 282 randomly selected households and semi-structured interviews carried out with purposefully selected key informants representing institutional stakeholders. The results of this study show that the stakeholder categories identified in the two wetland areas are government departments, statutory bodies, non-governmental organisations, research institutions, academia and the local people. Most of the institutional stakeholders (55.6%) and household respondents (81.7%) do not have knowledge of EO data application in wetland conditions monitoring. Chi-Square test results further show that the households’ knowledge on use of remotely sensed derived products and services does not vary with age (p > 0.05), marital status (p > 0.05), level of education (p > 0.05), except gender in Driefontein (p < 0.05). However, the various stakeholder categories recommended wider application of earth observation products and services in the assessment of land use land cover changes, water quantity and quality, flood monitoring, fire outbreaks and the establishment of a wetland inventory to abate wetland degradation. The utilisation of EO data is constrained by lack of knowledge on the use of remotely sensed data in wetland ecological conditions assessment. Investment in capacity building initiatives for both technical and non-technical stakeholders involved in wetland management in Zimbabwe is required so as to enhance their skills in EO data utilisation.

Use of High-Resolution Land Cover Maps to Support the Maintenance of the NWI Geospatial Dataset: A Case Study in a Coastal New Orleans Region

The National Wetlands Inventory (NWI) is the most comprehensive wetland geospatial dataset in the United States. However, it can be time-consuming and costly to maintain. This study introduces automated algorithms and methods to support NWI maintenance. Through a wall-to-wall comparison between NWI and Coastal Change Analysis Program (C-CAP) datasets, a pixel-level difference product was generated at 1 m resolution. Building upon this, supplementary attributes describing wetland changes were incorporated into each NWI polygon. Additionally, new water polygons were extracted from C-CAP data, and regional statistics regarding wetland changes were computed for HUC12 watersheds. The 1 m difference product can indicate specific wetland change locations, such as wetland loss to impervious surfaces, the gain of open water bodies from uplands, and the conversion of drier vegetated wetlands to open water. The supplementary attributes can indicate the amount and percentage of wetland loss or water regime change for NWI polygons. Extracted new water polygons can serve as preliminary materials for generating NWI standard-compliant products, expediating NWI maintenance processes while reducing costs. Regional statistics of wetland change can help target watersheds with the most significant changes for maintenance, thereby reducing work areas. The approaches we present hold significant value in supporting NWI maintenance.

The influence of roads on depressional storage capacity estimates from high-resolution LiDAR DEMs in a Canadian Prairie agricultural basin

The Canadian Prairies are a post-glacial agricultural landscape, where millions of small depressions store surface water, form wetlands and control runoff contributing area. Their management is key to flood and drought hydrology, groundwater recharge, ecological integrity, migratory bird habitat and agricultural productivity. Depression drainage and infilling is common in the region, where it is often used to increase cropped area. The regularly spaced, rural ‘grid-road‘ network also impedes drainage, but associated culvert drainage can mitigate those effects. Management of depressions can be informed by hydrological modelling, but accurate surface water storage capacity estimates are needed to ensure accurate model results. Simple representation of road embankments in digital elevation models (DEMs) neglects the effects of culvert drainage. Here, a raster-based depression-filling algorithm was used to delineate depressions from three LiDAR-derived DEMs: a 10-m DEM with roads intact, a 2-m DEM with roads intact, and a 2-m DEM with roads breached at culvert locations. Road breaching was conducted manually in the 2-m DEM to remove artifact depressions that form alongside roads where culverts exist. Results indicated that increasing DEM resolution from 10-m to 2-m in a 393.5 km2 basin did not significantly change depression area or storage capacity estimates; however, breaching roads in the 2-m DEM decreased depression area by 29% (from 98.5 km2 to 69.8 km2) and estimated storage capacity by 48% (from 47.4 × 106 m3 to 23.8 × 106 m3), compared to leaving roads intact in the 2-m DEM. Depressions delineated from the 2-m roads-breached DEM also covered 48% more area and offered 53% more storage capacity than Canadian Wetland Inventory (CWI) aerial-photograph delineated wetlands, which occupied 47.1 km2 with an estimated storage capacity of 15.5 × 106 m3. The implications of these results for the ability of hydrological models to calculate runoff contributing areas and streamflow are discussed.

Peruvian Wetlands: National Survey, Diagnosis, and Further Steps toward Their Protection

Wetlands are crucial hydrological features that provide benefits, including ecosystem services and habitat conservation, protection from flooding associated with sea level rise and extreme events, water storage and treatment, recreation, and aesthetics, among others. Peru is one of the few countries in Latin America that has not developed a national wetland inventory. While this can in part be attributed to the limited availability of peer-reviewed publications in the English literature, a larger quantity of Peruvian wetland-related publications are available in Spanish, and are less accessible to the global population. In this paper, we developed an extensive review and synthesis of the Spanish literature, including university theses, peer-reviewed articles, and government reports. Our report focuses on evaluating the state of the art of Peruvian wetlands in terms of temporal evolution, geographical distribution, vested institutions, research topics (e.g., water quality, fauna, flora, microorganisms, etc.), and advances toward the conservation of wetlands. The analysis identified 274 wetland-related publications in Spanish (188 theses, 83 peer-reviewed, and 6 government reports) and a temporal increase in dissemination over the past two decades. The reports encompassed 161 distinct wetlands distributed nationwide; however, most of the investigations focused on only a few wetland bodies with a disproportionate concentration in just three administrative regions. This reveals that wetland-related research is unevenly distributed in Peru, and highlights a need to extend this knowledge to underrepresented systems and regions. Although Peru ranks third in South America for protected (Ramsar) wetland areas, case studies have revealed that wetlands in the country are vulnerable to human activities. Recent national legislation established in 2021 should help to address this challenge, as before this time, there was a more decentralized approach, whereby each administrative region held responsibility for the protection of their own wetlands. Collective findings indicate that research activity should be increased nationally in order to better understand the function and benefits of wetlands throughout Peru, in addition to the continued development and enforcement of regulations designed to protect these valuable ecosystems. Finally, since a national Peruvian wetland inventory is urgently needed, this analysis provides a baseline for this development of, as well as identifying gaps in, knowledge needed for appropriate national representation.

Iranian wetland inventory map at a spatial resolution of 10 m using Sentinel-1 and Sentinel-2 data on the Google Earth Engine cloud computing platform

Detailed wetland inventories and information about the spatial arrangement and the extent of wetland types across the Earth's surface are crucially important for resource assessment and sustainable management. In addition, it is crucial to update these inventories due to the highly dynamic characteristics of the wetlands. Remote sensing technologies capturing high-resolution and multi-temporal views of landscapes are incredibly beneficial in wetland mapping compared to traditional methods. Taking advantage of the Google Earth Engine's computational power and multi-source earth observation data from Sentinel-1 multi-spectral sensor and Sentinel-2 radar, we generated a 10m nationwide wetlands inventory map for Iran. The whole country is mapped using an object-based image processing framework, containing SNIC superpixel segmentation and a Random Forest classifier that was performed for four different ecological zones of Iran separately. Reference data was provided by different sources and through both field and office-based methods. Almost 70% of this data was used for the training stage and the other 30% for evaluation. The whole map overall accuracy was 96.39% and the producer's accuracy for wetland classes ranged from nearly 65 to 99%. It is estimated that 22,384 km2 of Iran are covered with water bodies and wetland classes, and emergent and shrub-dominated are the most common wetland classes in Iran. Considering the water crisis that has been started in Iran, the resulting ever-demanding map of Iranian wetland sites offers remarkable information about wetland boundaries and spatial distribution of wetland species, and therefore it is helpful for both governmental and commercial sectors.

RadWet: An Improved and Transferable Mapping of Open Water and Inundated Vegetation Using Sentinel-1

Mapping the spatial and temporal dynamics of tropical herbaceous wetlands is vital for a wide range of applications. Inundated vegetation can account for over three-quarters of the total inundated area, yet widely used EO mapping approaches are limited to the detection of open water bodies. This paper presents a new wetland mapping approach, RadWet, that automatically defines open water and inundated vegetation training data using a novel mixture of radar, terrain, and optical imagery. Training data samples are then used to classify serial Sentinel-1 radar imagery using an ensemble machine learning classification routine, providing information on the spatial and temporal dynamics of inundation every 12 days at a resolution of 30 m. The approach was evaluated over the period 2017–2022, covering a range of conditions (dry season to wet season) for two sites: (1) the Barotseland Floodplain, Zambia (31,172 km2) and (2) the Upper Rupununi Wetlands in Guyana (11,745 km2). Good agreement was found at both sites using random stratified accuracy assessment data (n = 28,223) with a median overall accuracy of 89% in Barotseland and 80% in the Upper Rupununi, outperforming existing approaches. The results revealed fine-scale hydrological processes driving inundation patterns as well as temporal patterns in seasonal flood pulse timing and magnitude. Inundated vegetation dominated wet season wetland extent, accounting for a mean 80% of total inundation. RadWet offers a new way in which tropical wetlands can be routinely monitored and characterised. This can provide significant benefits for a range of application areas, including flood hazard management, wetland inventories, monitoring natural greenhouse gas emissions and disease vector control.

A Random Forest-Based Multi-Index Classification (RaFMIC) Approach to Mapping Three-Decadal Inundation Dynamics in Dryland Wetlands Using Google Earth Engine

Australian inland riparian wetlands located east of the Great Dividing Range exhibit unique, hydroecological characteristics. These flood-dependent aquatic systems located in water-limited regions are declining rapidly due to the competitive demand for water for human activities, as well as climate change and variability. However, there exist very few reliable data to characterize inundation change conditions and quantify the impacts of the loss and deterioration of wetlands. A long-term time record of wetland inundation maps can provide a crucial baseline to monitor, assess, and assist the management and conservation of wetland ecosystems. This study presents a random forest-based multi-index classification algorithm (RaFMIC) on the Google Earth Engine (GEE) platform to efficiently construct a temporally dense, three-decadal time record of inundation maps of the southeast Australian riparian inland wetlands. The method was tested over the Macquarie Marshes located in the semiarid region of NSW, Australia. The results showed a good accuracy when compared against high-spatial resolution imagery. The total inundated area was consistent with precipitation and streamflow patterns, and the temporal dynamics of vegetation showed good agreement with the inundation maps. The inundation time record was analysed to generate inundation probability maps, which were in a good agreement with frequently flooded areas simulated by a hydrodynamic model and the distribution of flood-dependent vegetation species. The long-term, time-dense inundation maps derived from the RaFMIC method can provide key information to assess the condition and health of wetland ecosystems and have the potential to improve wetland inventory with spatially explicit water regime information. RaFMIC can be adapted over other dryland wetlands, as an effective semiautomated method of mapping long-term inundation dynamics.

A satellite-based monitoring system for quantifying surface water and mesic vegetation dynamics in a semi-arid region

Semi-arid and arid systems cover one third of the earth’s land surface, and are becoming increasingly drier, but existing datasets do not capture all of the types of water resources that sustain these systems. In semi-arid environments, small surface water bodies and areas of mesic vegetation (wetlands, wet meadows, riparian zones) function as critical water resources. However, the most commonly-used maps of water resources are derived from the Landsat time series or single date aerial photographs, and are too coarse either spatially or temporally to effectively monitor water resource dynamics. In this study, we produced a Sentinel Fusion (SF) water resources product for a semi-arid mountainous region of the western United States, which includes monthly maps of both a) surface water and b) mesic vegetation at 10 m spatial resolution using freely available Earth observation data on an open access platform. We applied random forest classifiers to optical data from the Sentinel-2 time series, synthetic aperture radar (SAR) data from the Sentinel-1 time series, and topographic variables. We compared our SF product with three commonly used and publicly available datasets in the western U.S. We found that our surface water class contained fewer omission errors than a leading global surface water product (94 % producer’s accuracy (PA) vs 84 %) and comparable user’s accuracy (UA) (91 % vs 97 %) with commission errors occurring largely in mixed water pixels. Our mesic vegetation class had up to 43 % higher PAs compared to the National Wetlands Inventory (NWI) estimates and up to 78 % higher UAs over the Sage Grouse Initiative mesic resources maps during the most critical part of the water year. We found that while inclusion of SAR data from the C-band Sentinel-1 sensor consistently improved estimates of water resources in each of the last four months of the 2021 water year when compared to optical-only + topographic variables, only in September did those improvements lie outside of the 95 % confidence interval. With nine times finer spatial resolution and more frequent image collection, our SF maps characterize intra-annual dynamics of smaller water bodies (<30 m wide) and mesic vegetation integral to ecosystem functioning in semi-arid systems compared to leading Landsat-derived products. Further, our workflow is easily reproducible using freely available data on an open access platform, and can be adopted to help guide land use decisions related to water resources by farmers, ranchers, and conservationists in semi-arid environments.