An official website of the United States government
Abstract. The coastal area of New Hanover County in North Carolina encompasses diverse wetland habitats influenced by unique coastal and tidal dynamics, with researchers examining the impacts of landscape changes, sea-level rise, and climate fluctuations on wetland health and biodiversity. This study integrates multispectral imagery data, LiDAR, and additional sources to enhance classification accuracy. The study also addresses binary classification for wetland and non-wetland classification and a multi-classification for different wetland classes, leveraging on the Random Forest algorithm which significantly improved the overall accuracy of wetland mapping. The Random Forest model’s performance in different scenarios was evaluated, with Scenario 1 achieving an overall accuracy of nearly 93.9%, Scenario 2 achieving an overall accuracy of 93.5%, Scenario 3 achieving an overall accuracy of 94.1%, and Scenario 4 achieving an overall accuracy of 88.2%. These results underscore the model’s effectiveness in accurately classifying coastal wetland areas under diverse remote sensing scenarios, highlighting its potential for practical applications in wetland mapping and ecological research.
more »
« less
Predicting nitrate exposure from groundwater wells using machine learning and meteorological conditions
Abstract Private groundwater wells can be unmonitored sources of contaminated water that can harm human health. Developing models that predict exposure could allow residents to take action to reduce risk. Machine learning models have been successful in predicting nitrate contamination using geospatial information such as proximity to nitrate sources, but previous models have not considered meteorological factors that change temporally. In this study, we test random forest (regression and classification) and linear regression models to predict nitrate contamination using rainfall, temperature, and readily available soil parameters. We trained and tested models for (1) all of North Carolina, (2) each geographic region in North Carolina, (3) a three‐county region with a high density of animal agriculture, and (4) a three‐county region with a low density of animal agriculture. All regression models had poor predictive performance (R2 < 0.09). The random forest classification model for the coastal plain showed fair agreement (Cohen'sκ = 0.23) when trying to predict whether contamination occurred. All other classification models had slight or poor predictive performance. Our results show that temporal changes in rainfall and temperature, or in combination with soil data, are not enough to predict nitrate contamination in most areas of North Carolina. The low level of contamination (<25%) measured during the study could have contributed to the poor performance of the models.
more »
« less
- Award ID(s):
- 2121788
- PAR ID:
- 10476385
- Publisher / Repository:
- Wiley-Blackwell
- Date Published:
- Journal Name:
- JAWRA Journal of the American Water Resources Association
- Volume:
- 60
- Issue:
- 2
- ISSN:
- 1093-474X
- Format(s):
- Medium: X Size: p. 639-651
- Size(s):
- p. 639-651
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Multiple species of ticks, including Ixodes scapularis (Say, Ixodida:Ixodidae), Amblyomma americanum (L., Ixodida:Ixodidae), and Dermacentor variabilis (Say, Ixodida:Ixodidae), occur in high and increasing abundance in both the northeast and southeast United States. North Carolina is at the nexus of spread of these species, with high occurrence and abundance of I. scapularis to the north and A. americanum to the south. Despite this, there are few records of these species in the Piedmont of North Carolina, including the greater Charlotte metropolitan area. Here, we update the known occurrence and abundance of these species in the North Carolina Piedmont. We surveyed for ticks using cloth drags, CO2 traps, and leaf litter samples at a total of 79 sites within five locations: Mecklenburg County, South Mountains State Park, Stone Mountain State Park, Duke Forest, and Morrow Mountain State Park, all in North Carolina, during the late spring, summer, and fall seasons of 2019. From these surveys, we had only 20 tick captures, illuminating the surprisingly low abundance of ticks in this region of North Carolina. Our results indicate the possibility of underlying habitat and host factors limiting tick distribution and abundance in the North Carolina Piedmont.more » « less
-
Abstract AimEfforts to predict the responses of soil fungal communities to climate change are hindered by limited information on how fungal niches are distributed across environmental hyperspace. We predict the climate sensitivity of North American soil fungal assemblage composition by modelling the ecological niches of several thousand fungal species. LocationOne hundred and thirteen sites in the United States and Canada spanning all biomes except tropical rain forest. Major Taxa StudiedFungi. Time Period2011–2018. MethodsWe combine internal transcribed spacer (ITS) sequences from two continental‐scale sampling networks in North America and cluster them into operational taxonomic units (OTUs) at 97% similarity. Using climate and soil data, we fit ecological niche models (ENMs) based on logistic ridge regression for all OTUs present in at least 10 sites (n = 8597). To describe the compositional turnover of soil fungal assemblages over climatic gradients, we introduce a novel niche‐based metric of climate sensitivity, the Sørensen climate sensitivity index. Finally, we map climate sensitivity across North America. ResultsENMs have a mean out‐of‐sample predictive accuracy of 73.8%, with temperature variables being strong predictors of fungal distributions. Soil fungal climate niches clump together across environmental space, which suggests common physiological limits and predicts abrupt changes in composition with respect to changes in climate. Soil fungi in North American climates are more likely to be limited by cold and dry conditions than by warm and wet conditions, and ectomycorrhizal fungi generally tolerate colder temperatures than saprotrophic fungi. Sørensen climate sensitivity exhibits a multimodal distribution across environmental space, with a peak in climates corresponding to boreal forests. Main ConclusionsThe boreal forest occupies an especially precarious region of environmental space for the composition of soil fungal assemblages in North America, as even small degrees of warming could trigger large compositional changes characterized mainly by an influx of warm‐adapted species.more » « less
-
null (Ed.)Understanding the dominant soil nitrogen (N) cycling processes in southern Appalachian forests is crucial for predicting ecosystem responses to changing N deposition and climate. The role of anaerobic nitrogen cycling processes in well-aerated soils has long been questioned, and recent N cycling research suggests it needs to be re-evaluated. We assessed gross and potential rates of soil N cycling processes, including mineralization, nitrification, denitrification, nitrifier denitrification, and dissimilatory nitrate reduction to ammonium (DNRA) in sites representing a vegetation and elevation gradient in the U.S. Department of Agriculture (USDA) Forest Service Experimental Forest, Coweeta Hydrologic Laboratory in southwestern North Carolina, USA. N cycling processes varied among sites, with gross mineralization and nitrification being greatest in high-elevation northern hardwood forests. Gaseous N losses via nitrifier denitrification were common in all ecosystems but were greatest in northern hardwood. Ecosystem N retention via DNRA (nitrification-produced NO3 reduced to NH4) ranged from 2% to 20% of the total nitrification and was highest in the mixed-oak forest. Our results suggest the potential for gaseous N losses through anaerobic processes (nitrifier denitrification) are prevalent in well-aerated forest soils and may play a key role in ecosystem N cycling.more » « less
-
Abstract Predicting the edges of species distributions is fundamental for species conservation, ecosystem services, and management decisions. In North America, the location of the upstream limit of fish in forested streams receives special attention, because fish-bearing portions of streams have more protections during forest management activities than fishless portions. We present a novel model development and evaluation framework, wherein we compare 26 models to predict upper distribution limits of trout in streams. The models used machine learning, logistic regression, and a sophisticated nested spatial cross-validation routine to evaluate predictive performance while accounting for spatial autocorrelation. The model resulting in the best predictive performance, termed UPstream Regional LiDAR Model for Extent of Trout (UPRLIMET), is a two-stage model that uses a logistic regression algorithm calibrated to observations of Coastal Cutthroat Trout ( Oncorhynchus clarkii clarkii ) occurrence and variables representing hydro-topographic characteristics of the landscape. We predict trout presence along reaches throughout a stream network, and include a stopping rule to identify a discrete upper limit point above which all stream reaches are classified as fishless. Although there is no simple explanation for the upper distribution limit identified in UPRLIMET, four factors, including upstream channel length above the point of uppermost fish, drainage area, slope, and elevation, had highest importance. Across our study region of western Oregon, we found that more of the fish-bearing network is on private lands than on state, US Bureau of Land Mangement (BLM), or USDA Forest Service (USFS) lands, highlighting the importance of using spatially consistent maps across a region and working across land ownerships. Our research underscores the value of using occurrence data to develop simple, but powerful, prediction tools to capture complex ecological processes that contribute to distribution limits of species.more » « less
