A generalized machine learning approach for dissolved oxygen estimation at multiple spatiotemporal scales using remote sensing.
TL;DR: In this paper, support vector regression (SVR) models were developed and validated using the remote sensing reflectance derived from both Landsat and Moderate Resolution Imaging Spectroradiometer (MODIS) data and synchronous DO measurements and water temperature of Lake Huron and three other inland waterbodies.
About: This article is published in Environmental Pollution.The article was published on 2021-11-01. It has received 15 citations till now. The article focuses on the topics: Moderate-resolution imaging spectroradiometer.
Citations
More filters
TL;DR: In this paper , a multimodal deep learning (MDL) model was developed and rigorously validated using atmospherically corrected Landsat remote sensing reflectance data and synchronous water quality measurements for estimating long-term Chlorophyll-a (Chl-a ), total phosphorus (TP), and total nitrogen (TN) in Lake Simcoe, Canada.
Abstract: Remote sensing provides full-coverage and dynamic water quality monitoring with high efficiency and low consumption. Deep learning (DL) has been progressively used in water quality retrieval because it efficiently captures the potential relationship between target variables and imagery. In this study, the multimodal deep learning (MDL) models were developed and rigorously validated using atmospherically corrected Landsat remote sensing reflectance data and synchronous water quality measurements for estimating long-term Chlorophyll- a (Chl- a ), total phosphorus (TP), and total nitrogen (TN) in Lake Simcoe, Canada. Since TP and TN are non-optically active, their retrievals were based on the fact that they are closely related to the optically active constituents (OACs) such as Chl- a . We trained the MDL models with one in-situ measured data set (for Chl- a , N = 315, for TP and TN, N = 303), validated the models with two independent data sets ( N = 147), and compared the model performances with several DL, machine learning, and empirical algorithms. The results indicated that the MDL models adequately estimated Chl- a (mean absolute error (MAE) = 32.57%, Bias = 10.61%), TP (MAE = 42.58%, Bias = −2.82%), and TN (MAE = 35.05%, Bias = 13.66%), and outperformed several other candidate algorithms, namely the progressively decreasing deep neural network (DNN), a DNN with trainable parameters similar to MDL but without splitting input features, the eXtreme Gradient Boosting, the support vector regression, the NASA Ocean Color two-band and three-band ratio algorithms, and another empirical algorithm of Landsat data in clear lakes. Using the MDL models, we reconstructed the historical spatiotemporal patterns of Chl- a , TP, and TN in Lake Simcoe since 1984, and investigated the effects of two water quality improvement programs. In addition, the physical mechanism and interpretability of the MDL models were explored by quantifying the contribution of each feature to the model outputs. The framework proposed in this study provides a practical method for long-term Chl- a , TP, and TN estimation at the regional scale.
13 citations
TL;DR: In this paper, support vector regression (SVR) was employed for prediction of the model in both standalone and hybrid forms, which consisted in SVR combined with metaheuristic algorithms of chicken swarm optimization (CSO), social ski-driver (SSD) optimization, Black widow optimization (BWO), and the Algorithm of the innovative gunner (AIG).
Abstract: Water is one of the most essential elements in nature that forms the basis of human life and contributes to the economic growth and development of societies. Safe water is closely related to environmental health and activities. The lives of all the animals on our planet depend on water and oxygen. Moreover, sufficient dissolved oxygen (DO) is crucial for the survival of aquatic animals. In the present research, temperature (T) and flow (Q) variables were used to predict DO. The time series were monthly and data were related to the Cumberland River in the southern United States from 2008 to 2018. Support vector regression (SVR) was employed for prediction of the model in both standalone and hybrid forms. The employed hybrid models consisted in SVR combined with metaheuristic algorithms of chicken swarm optimization (CSO), social ski-driver (SSD) optimization, Black widow optimization (BWO), and the Algorithm of the innovative gunner (AIG). Pearson correlation coefficient was utilized to select the best input combination. Box plots and Taylor diagrams were employed in the interpretation of the results. It was observed that all the four hybrid models achieved better results. Also according to the evaluation criteria among the models used the following were found: SVR–AIG with the coefficient of determination (R2 = 0.963), the root mean square error (RMSE = 0.644 mg/l), the mean absolute value of error (MAE = 0.568 mg/l), the Nash–Sutcliffe coefficient (NS = 0.864), and bias percentage (BIAS = 0.001). Overall the research showed that hybrid models increased the accuracy of the single SVR model by 6.52–1.75%.
12 citations
TL;DR: In this article , the authors employed GA for optimization of the selection and layout of low impact development (LID) facilities and layouts for a sub-catchment is important for designing stormwater management system.
12 citations
TL;DR: Long-term continuous monitoring (LTCM) of water quality can bring far-reaching influences on water ecosystems by providing spatiotemporal data sets of diverse parameters and enabling operation of water and wastewater treatment processes in an energy-saving and cost-effective manner as discussed by the authors .
Abstract: Long-term continuous monitoring (LTCM) of water quality can bring far-reaching influences on water ecosystems by providing spatiotemporal data sets of diverse parameters and enabling operation of water and wastewater treatment processes in an energy-saving and cost-effective manner. However, current water monitoring technologies are deficient for long-term accuracy in data collection and processing capability. Inadequate LTCM data impedes water quality assessment and hinders the stakeholders and decision makers from foreseeing emerging problems and executing efficient control methodologies. To tackle this challenge, this review provides a forward-looking roadmap highlighting vital innovations toward LTCM, and elaborates on the impacts of LTCM through a three-hierarchy perspective: data, parameters, and systems. First, we demonstrate the critical needs and challenges of LTCM in natural resource water, drinking water, and wastewater systems, and differentiate LTCM from existing short-term and discrete monitoring techniques. We then elucidate three steps to achieve LTCM in water systems, consisting of data acquisition (water sensors), data processing (machine learning algorithms), and data application (with modeling and process control as two examples). Finally, we explore future opportunities of LTCM in four key domains, water, energy, sensing, and data, and underscore strategies to transfer scientific discoveries to general end-users.
8 citations
References
More filters
TL;DR: High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated and the performance of the support- vector network is compared to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
Abstract: The support-vector network is a new learning machine for two-group classification problems. The machine conceptually implements the following idea: input vectors are non-linearly mapped to a very high-dimension feature space. In this feature space a linear decision surface is constructed. Special properties of the decision surface ensures high generalization ability of the learning machine. The idea behind the support-vector network was previously implemented for the restricted case where the training data can be separated without errors. We here extend this result to non-separable training data.
High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated. We also compare the performance of the support-vector network to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
37,861 citations
8,699 citations
TL;DR: This paper reviews remote sensing implementations of support vector machines (SVMs), a promising machine learning methodology that is particularly appealing in the remote sensing field due to their ability to generalize well even with limited training samples.
Abstract: A wide range of methods for analysis of airborne- and satellite-derived imagery continues to be proposed and assessed. In this paper, we review remote sensing implementations of support vector machines (SVMs), a promising machine learning methodology. This review is timely due to the exponentially increasing number of works published in recent years. SVMs are particularly appealing in the remote sensing field due to their ability to generalize well even with limited training samples, a common limitation for remote sensing applications. However, they also suffer from parameter assignment issues that can significantly affect obtained results. A summary of empirical results is provided for various applications of over one hundred published works (as of April, 2010). It is our hope that this survey will provide guidelines for future applications of SVMs and possible areas of algorithm enhancement.
2,546 citations
TL;DR: A review of the current status of selected remote sensing algorithms for estimating land surface temperature from thermal infrared (TIR) data is presented in this article, along with a survey of the algorithms employed for obtaining LST from space-based TIR measurements.
1,470 citations
Smithsonian Environmental Research Center1, University of California, San Diego2, Leibniz Institute of Marine Sciences3, University of Liège4, Monterey Bay Aquarium Research Institute5, Lund University6, Centre national de la recherche scientifique7, Fisheries and Oceans Canada8, Cayetano Heredia University9, University of the Philippines Diliman10, State University of New York College of Environmental Science and Forestry11, Kuwait Institute for Scientific Research12, University of Cape Town13, Department of Agriculture, Forestry and Fisheries14, Louisiana State University15, University of Maryland Center for Environmental Science16, University of South Florida St. Petersburg17, Polish Academy of Sciences18, University of Hong Kong19, East China Normal University20
TL;DR: Improved numerical models of oceanographic processes that control oxygen depletion and the large-scale influence of altered biogeochemical cycles are needed to better predict the magnitude and spatial patterns of deoxygenation in the open ocean, as well as feedbacks to climate.
Abstract: BACKGROUND Oxygen concentrations in both the open ocean and coastal waters have been declining since at least the middle of the 20th century. This oxygen loss, or deoxygenation, is one of the most important changes occurring in an ocean increasingly modified by human activities that have raised temperatures, CO 2 levels, and nutrient inputs and have altered the abundances and distributions of marine species. Oxygen is fundamental to biological and biogeochemical processes in the ocean. Its decline can cause major changes in ocean productivity, biodiversity, and biogeochemical cycles. Analyses of direct measurements at sites around the world indicate that oxygen-minimum zones in the open ocean have expanded by several million square kilometers and that hundreds of coastal sites now have oxygen concentrations low enough to limit the distribution and abundance of animal populations and alter the cycling of important nutrients. ADVANCES In the open ocean, global warming, which is primarily caused by increased greenhouse gas emissions, is considered the primary cause of ongoing deoxygenation. Numerical models project further oxygen declines during the 21st century, even with ambitious emission reductions. Rising global temperatures decrease oxygen solubility in water, increase the rate of oxygen consumption via respiration, and are predicted to reduce the introduction of oxygen from the atmosphere and surface waters into the ocean interior by increasing stratification and weakening ocean overturning circulation. In estuaries and other coastal systems strongly influenced by their watershed, oxygen declines have been caused by increased loadings of nutrients (nitrogen and phosphorus) and organic matter, primarily from agriculture; sewage; and the combustion of fossil fuels. In many regions, further increases in nitrogen discharges to coastal waters are projected as human populations and agricultural production rise. Climate change exacerbates oxygen decline in coastal systems through similar mechanisms as those in the open ocean, as well as by increasing nutrient delivery from watersheds that will experience increased precipitation. Expansion of low-oxygen zones can increase production of N 2 O, a potent greenhouse gas; reduce eukaryote biodiversity; alter the structure of food webs; and negatively affect food security and livelihoods. Both acidification and increasing temperature are mechanistically linked with the process of deoxygenation and combine with low-oxygen conditions to affect biogeochemical, physiological, and ecological processes. However, an important paradox to consider in predicting large-scale effects of future deoxygenation is that high levels of productivity in nutrient-enriched coastal systems and upwelling areas associated with oxygen-minimum zones also support some of the world’s most prolific fisheries. OUTLOOK Major advances have been made toward understanding patterns, drivers, and consequences of ocean deoxygenation, but there is a need to improve predictions at large spatial and temporal scales important to ecosystem services provided by the ocean. Improved numerical models of oceanographic processes that control oxygen depletion and the large-scale influence of altered biogeochemical cycles are needed to better predict the magnitude and spatial patterns of deoxygenation in the open ocean, as well as feedbacks to climate. Developing and verifying the next generation of these models will require increased in situ observations and improved mechanistic understanding on a variety of scales. Models useful for managing nutrient loads can simulate oxygen loss in coastal waters with some skill, but their ability to project future oxygen loss is often hampered by insufficient data and climate model projections on drivers at appropriate temporal and spatial scales. Predicting deoxygenation-induced changes in ecosystem services and human welfare requires scaling effects that are measured on individual organisms to populations, food webs, and fisheries stocks; considering combined effects of deoxygenation and other ocean stressors; and placing an increased research emphasis on developing nations. Reducing the impacts of other stressors may provide some protection to species negatively affected by low-oxygen conditions. Ultimately, though, limiting deoxygenation and its negative effects will necessitate a substantial global decrease in greenhouse gas emissions, as well as reductions in nutrient discharges to coastal waters.
1,469 citations