scholarly journals Multispectral remote sensing inversion for city landscape water eutrophication based on Genetic Algorithm-Support Vector Machine

2014 ◽  
Vol 49 (3) ◽  
pp. 285-293 ◽  
Author(s):  
Aidi Huo ◽  
Jia Zhang ◽  
Changlu Qiao ◽  
Chenlong Li ◽  
Juan Xie ◽  
...  
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Water Quality ◽  
Urban Landscape ◽  
In Situ Monitoring ◽  
Support Vector ◽  
Landscape Water ◽  
Water Eutrophication ◽  
Urban Landscape Water

Eutrophication has become the primary water quality issue for many urban landscape waters in the world. It is a focus in this paper which analyzes Enhanced Thematic Mapper images and quality observation data for 12 consecutive years in 20 parts of the urban landscape water in Xi'an City, China. A water quality model for urban landscape water based on Support Vector Machine (SVM) was established. Based on in situ monitoring data, the model is compared with water quality retrieving methods of multiple regression and back propagation neural network. Results show that the Genetic Algorithm-SVM (GA-SVM) method has better prediction accuracy than the inversion results of the neural network and the traditional statistical regression method. In short, GA-SVM provides a new method for remote sensing monitoring of urban water eutrophication and has more accurate predictions in inversion results [such as chlorophyll a (Chl-a)] in the Xi'an area. Additionally, remote sensing results highly agreed with in situ monitoring data, indicating that the technology is effective and less costly than in situ monitoring. The technology also can be used to evaluate large lake eutrophication.

Quantifying Urban Landscape Water Conservation Potential Using High Resolution Remote Sensing and GIS

2011 ◽  
Vol 77 (11) ◽  
pp. 1113-1122 ◽  
Author(s):  
Fayek A. Farag ◽  
Christopher M.U. Neale ◽  
Roger K. Kjelgren ◽  
Joanna Endter-Wada
Keyword(s):  
Remote Sensing ◽  
High Resolution ◽  
Water Conservation ◽  
Urban Landscape ◽  
Remote Sensing And Gis ◽  
Landscape Water ◽  
Urban Landscape Water

scholarly journals Geospatial assessment of the Mokra Sura river ecological condition using remote sensing and in situ monitoring data

2019 ◽  
Vol 27 (3) ◽  
pp. 422-430
Author(s):  
Mykola М. Kharytonov ◽  
Andriy М. Pugach ◽  
Sergey А. Stankevich ◽  
Anna O. Кozlova
Keyword(s):  
Remote Sensing ◽  
Water Quality ◽  
Surface Water ◽  
River Basin ◽  
Ecological Status ◽  
The State ◽  
Water Bodies ◽  
In Situ Monitoring ◽  
Treated Wastewater

The use of remote sensing methods for environmental monitoring of the surface water quality is proved. Regression relationships are consistent with ground-based measurements at sampling sites in water bodies and are an effective tool for assessing the ecological status of water bodies. The state of the water bodies of the Mokra Sura river basin varies considerably. The best is the water quality in the upper part of the Mokra Sura river, the worst – in the middle and lower parts. The factors of water pollution are discharges of not enough treated wastewater of industrial enterprises of the Kamyans’koy and Dniprovs’koy industrial agglomeration. The purpose of our search included the following tasks: (a) calculation of integrated environmental water quality indices; b) obtaining satellite information, processing of multispectral satellite images of water bodies using appropriate applied software techniques; c) establishment of statistical dependencies between water quality indexes obtained for biotopically space images and data of actual in situ measurements. The results of systematic hydrochemical control of the Mokra Sura river basin from 2007 to 2011 years were initial data in 4 control areas located in the Dnipropetrovsk region: 1 – the Sursko-Litovske village; 2 – the Bratske village; 3 – the Novomykolayvka village; 4 – the Novooleksandryvka village. Environmental assessment of the water quality of the Mokra Sura river within the Dnipropetrovsk region was based on the calculation the integrated environmental index ( IEI ). Priority pollutants in this case are oil products and ions 2−SO 4, 2 + Mg , 2 + Zn , 6 + Cr . Two images with a difference in three years in April 2015 and May 2017 were used to determine the current changes in the land cover of the study area. Geomorphological assessment of the water network of the Morka Sura river was performed using satellite radar interferometry. Multispectral images of Landsat 5/TM (2007-2011) and Sentinel 2B/MSI (2017) satellite systems were used forremote assessment of water bodies in the study area of the Mokra Sura river basin. The multispectral index TCW (Tasseled Cap Wetness) was used to measure the spectral reflection of the aquatic environment along of the Mokra Sura river flow. The main advantage of the studies is a demonstration of remote sensing capabilities to estimate Mokra Sura river ecological status not only in individual sites, but also throughout the flow – from source to mouth. Follow the necessity to use water from the Mokra Sura river for irrigation, the level of soil water erosion can only increase and enhance the negative processes of eutrophication of reservoirs. Long term technogenic pollution requires information about the state of surface water of fishery, drinking and municipal water use facilities as an integral part of the aquatic ecosystem, the habitat of aquatic organisms and as a resource of drinking water supply. Over 80% of the Mokra Sura river basin surface (IEI 4-12) belong to the classes with the assessment of dirty, very and extremely dirty. The results of studies using remote sensing indicate the need to reduce the streams of not enough treated wastewater to the the Mokra Sura river. The obtained data can be used for ecological assessment of the current and retrospective state of water bodies, development of forecasts of rivers pollution.

Remote sensing inversion of water quality in coastal sea area based on machine learning: a case study of Shenzhen bay, China

2021 ◽  
Author(s):  
Xiaotong Zhu ◽  
Jinhui Jeanne Huang
Keyword(s):  
Machine Learning ◽  
Remote Sensing ◽  
Water Quality ◽  
Predictive Accuracy ◽  
Water Environment ◽  
Quality Parameters ◽  
Machine Learning Algorithms ◽  
Dynamic Monitoring ◽  
Support Vector ◽  
Seawater Quality

<p>Remote sensing monitoring has the characteristics of wide monitoring range, celerity, low cost for long-term dynamic monitoring of water environment. With the flourish of artificial intelligence, machine learning has enabled remote sensing inversion of seawater quality to achieve higher prediction accuracy. However, due to the physicochemical property of the water quality parameters, the performance of algorithms differs a lot. In order to improve the predictive accuracy of seawater quality parameters, we proposed a technical framework to identify the optimal machine learning algorithms using Sentinel-2 satellite and in-situ seawater sample data. In the study, we select three algorithms, i.e. support vector regression (SVR), XGBoost and deep learning (DL), and four seawater quality parameters, i.e. dissolved oxygen (DO), total dissolved solids (TDS), turbidity(TUR) and chlorophyll-a (Chla). The results show that SVR is a more precise algorithm to inverse DO (R<sup>2</sup> = 0.81). XGBoost has the best accuracy for Chla and Tur inversion (R<sup>2</sup> = 0.75 and 0.78 respectively) while DL performs better in TDS (R<sup>2</sup> =0.789). Overall, this research provides a theoretical support for high precision remote sensing inversion of offshore seawater quality parameters based on machine learning.</p>

scholarly journals Method for Mapping Rice Fields in Complex Landscape Areas Based on Pre-Trained Convolutional Neural Network from HJ-1 A/B Data

2018 ◽  
Vol 7 (11) ◽  
pp. 418 ◽  
Author(s):  
Tian Jiang ◽  
Xiangnan Liu ◽  
Ling Wu
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Time Series ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Vegetation Index ◽  
Southern China ◽  
Series Data ◽  
Support Vector ◽  
Complex Landscape

Accurate and timely information about rice planting areas is essential for crop yield estimation, global climate change and agricultural resource management. In this study, we present a novel pixel-level classification approach that uses convolutional neural network (CNN) model to extract the features of enhanced vegetation index (EVI) time series curve for classification. The goal is to explore the practicability of deep learning techniques for rice recognition in complex landscape regions, where rice is easily confused with the surroundings, by using mid-resolution remote sensing images. A transfer learning strategy is utilized to fine tune a pre-trained CNN model and obtain the temporal features of the EVI curve. Support vector machine (SVM), a traditional machine learning approach, is also implemented in the experiment. Finally, we evaluate the accuracy of the two models. Results show that our model performs better than SVM, with the overall accuracies being 93.60% and 91.05%, respectively. Therefore, this technique is appropriate for estimating rice planting areas in southern China on the basis of a pre-trained CNN model by using time series data. And more opportunity and potential can be found for crop classification by remote sensing and deep learning technique in the future study.

scholarly journals Temporal Variation of Chlorophyll-a Concentrations in Highly Dynamic Waters from Unattended Sensors and Remote Sensing Observations

Sensors ◽  
2018 ◽  
Vol 18 (8) ◽  
pp. 2699 ◽  
Author(s):  
Jian Li ◽  
Liqiao Tian ◽  
Qingjun Song ◽  
Zhaohua Sun ◽  
Hongjing Yu ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Water Quality ◽  
Chlorophyll Fluorescence ◽  
Chlorophyll A ◽  
Temporal Variations ◽  
Lake Area ◽  
Short Term ◽  
Chl A ◽  
Spatio Temporal

Monitoring of water quality changes in highly dynamic inland lakes is frequently impeded by insufficient spatial and temporal coverage, for both field surveys and remote sensing methods. To track short-term variations of chlorophyll fluorescence and chlorophyll-a concentrations in Poyang Lake, the largest freshwater lake in China, high-frequency, in-situ, measurements were collected from two fixed stations. The K-mean clustering method was also applied to identify clusters with similar spatio-temporal variations, using remote sensing Chl-a data products from the MERIS satellite, taken from 2003 to 2012. Four lake area classes were obtained with distinct spatio-temporal patterns, two of which were selected for in situ measurement. Distinct daily periodic variations were observed, with peaks at approximately 3:00 PM and troughs at night or early morning. Short-term variations of chlorophyll fluorescence and Chl-a levels were revealed, with a maximum intra-diurnal ratio of 5.1 and inter-diurnal ratio of 7.4, respectively. Using geostatistical analysis, the temporal range of chlorophyll fluorescence and corresponding Chl-a variations was determined to be 9.6 h, which indicates that there is a temporal discrepancy between Chl-a variations and the sampling frequency of current satellite missions. An analysis of the optimal sampling strategies demonstrated that the influence of the sampling time on the mean Chl-a concentrations observed was higher than 25%, and the uncertainty of any single Terra/MODIS or Aqua/MODIS observation was approximately 15%. Therefore, sampling twice a day is essential to resolve Chl-a variations with a bias level of 10% or less. The results highlight short-term variations of critical water quality parameters in freshwater, and they help identify specific design requirements for geostationary earth observation missions, so that they can better address the challenges of monitoring complex coastal and inland environments around the world.

Sign in / Sign up

Export Citation Format

Share Document