Estimating Evapotranspiration in a Post-Fire Environment Using Remote Sensing and Machine Learning

In the hydrological cycle, evapotranspiration (ET) transfers moisture from the land surface to the atmosphere and is sensitive to disturbances such as wildfires. Ground-based pre- and post-fire measurements of ET are often unavailable, limiting the potential to understand the extent of wildfire impacts on the hydrological cycle. This research estimated both pre- and post-fire ET using remotely sensed variables and support vector machine (SVM) methods. Input variables (land surface temperature, modified soil-adjusted vegetation index, normalized difference moisture index, normalized burn ratio, precipitation, potential evapotranspiration, albedo and vegetation types) were used to train and develop 56 combinations that yielded 33 unique SVM models to predict actual ET. The models were trained to predict a spatial ET, the Operational Simplified Surface Energy Balance (SSEBop), for the 2003 Coyote Fire in San Diego, California (USA). The optimal SVM model, SVM-ET6, required six input variables and predicted ET for fifteen years with a root-mean-square error (RMSE) of 8.43 mm/month and a R2 of 0.89. The developed model was transferred and applied to the 2003 Old Fire in San Bernardino, California (USA), where a watershed balance approach was used to validate SVM-ET6 predictions. The annual water balance for ten out of fifteen years was within ±20% of the predicted values. This work demonstrated machine learning as a viable method to create a remotely-sensed estimate with wide applicability for regions with sparse data observations and information. This innovative work demonstrated the potential benefit for land and forest managers to understand and analyze the hydrological cycle of watersheds that experience acute disturbances based on this developed predictive ET model.

Download Full-text

mosquitoes.pdf

10.36227/techrxiv.11086871.v2 ◽

2019 ◽

Author(s):

Oladimeji Mudele ◽

Fabio M. Bayer ◽

Lucas Zanandrez ◽

Alvaro Eiras ◽

Paolo Gamba

Keyword(s):

Machine Learning ◽

Statistical Models ◽

Land Surface ◽

Vegetation Index ◽

Machine Learning Techniques ◽

Distribution Model ◽

Support Vector ◽

World Population ◽

Vector Population ◽

Learning Techniques

<div>Over 50% of the world population is at risk of mosquito-borne diseases. Female Ae. aegypti mosquito species transmit Zika, Dengue, and Chikungunya. The spread of these diseases correlate positively with the vector population, and this population depends on biotic and abiotic environmental factors including temperature, vegetation condition, humidity and precipitation. To combat virus outbreaks, information about vector population is required. To this aim, Earth observation (EO) data provide fast, efficient and economically viable means to estimate environmental features of interest. In this work, we present a temporal distribution model for adult female Ae. aegypti mosquitoes based on the joint use of the Normalized Difference Vegetation Index, the Normalized Difference Water Index, the Land Surface Temperature (both at day and night time), along with the precipitation information, extracted from EO data. The model was applied separately to data obtained during three different vector control and field data collection condition regimes, and used to explain the differences in environmental variable contributions across these regimes. To this aim, a random forest (RF) regression technique and its nonlinear features importance ranking based on mean decrease impurity (MDI) were implemented. To prove the robustness of the proposed model, other machine learning techniques, including support vector regression, decision trees and k-nearest neighbor regression, as well as artificial neural networks, and statistical models such as the linear regression model and generalized linear model were also considered. Our results show that machine learning techniques perform better than linear statistical models for the task at hand, and RF performs best. By ranking the importance of all features based on MDI in RF and selecting the subset comprising the most</div>

Download Full-text

Applying GRACE satellite data and hydro-meteorological parameters for predicting the combined terrestrial evapotranspiration index (CTEI) based on an improved SVM and a greedy regression approach

10.21203/rs.3.rs-222279/v1 ◽

2021 ◽

Author(s):

Ahmed Elbeltagi ◽

Mehdi Jamei ◽

Ankur Srivast ◽

Iman Ahmadianfar ◽

Mahfuzur Rahman ◽

...

Keyword(s):

Land Surface ◽

Crop Yields ◽

Potential Evapotranspiration ◽

Feedback Mechanism ◽

Climate Feedback ◽

Support Vector ◽

Linear Regression Method ◽

Terrestrial Water ◽

Input Variables ◽

Land Data Assimilation System

Abstract Drought has become relatively one of the widespread extreme events that affect socio-economic sphere, ecological balance, environmental conditions, crop-yields, climate feedback mechanism, and water resources management. Monitoring of the drought is one of the challenging tasks, hence required to develop new techniques in prediction of this natural calamity accurately and timely. In this work, Combined Terrestrial Evapotranspiration Index (CTEI) that is forced through precipitation (P) and potential evapotranspiration (PET) is used for estimating drought. Also, this is a novel index developed for Indus-Ganga river basin, which also uses a Gravity Recovery and Climate Experiment (GRACE) terrestrial water storage anomalies (TWSA) data. Thirteen input-variables comprised of GRACE-TWSA, Global Land Data Assimilation System GLDAS-TWSA, Groundwater Storage Anomaly (GWSA), P, Land Surface Temperature ( LST), Wind Speed (WS), Evapotranspiration (ET), Runoff, air temperature ( ), PET, net-shortwave (SWN) and long-wave (LWN) along with net radiation (NR) radiations were implemented for predicting CTEI. Further, thirteen different combinations were analyzed using an improved Sequential Minimal Optimization (SMO) algorithm for support vector machine (SVM) and a greedy linear regression method. Therefore, the objectives of this study are to develop several combinations based on machine learning models used for modeling CTEI, compare the accuracy and stability of these models in CTEI prediction, and select the best outcomes have been provided from these models. The performed error measures and statistical indicators prove that Combo 2 and 3 owing to SVM model are superior to the best combinations of Greedy model. Overall based on the results, the improved SVM is superior to Greedy model in estimating CTEI. SVM (Combo2: R=0.81) in spite of having a higher range of relative error, is superior to Greedy model (Combo3, R=0.77) for estimating CTEI. This method can be one of promising alternatives for estimating CTEI over major river basins across the world.

Download Full-text

mosquitoes.pdf

10.36227/techrxiv.11086871.v1 ◽

2019 ◽

Author(s):

Oladimeji Mudele ◽

Fabio M. Bayer ◽

Lucas Zanandrez ◽

Alvaro Eiras ◽

Paolo Gamba

Keyword(s):

Machine Learning ◽

Statistical Models ◽

Land Surface ◽

Vegetation Index ◽

Machine Learning Techniques ◽

Distribution Model ◽

Support Vector ◽

World Population ◽

Vector Population ◽

Learning Techniques

Download Full-text

Quantification of the Environmental Impacts of Highway Construction Using Remote Sensing Approach

Remote Sensing ◽

10.3390/rs13071340 ◽

2021 ◽

Vol 13 (7) ◽

pp. 1340

Author(s):

Shuailong Feng ◽

Shuguang Liu ◽

Lei Jing ◽

Yu Zhu ◽

Wende Yan ◽

...

Keyword(s):

Remote Sensing ◽

Environmental Impacts ◽

Land Surface ◽

Vegetation Index ◽

Highway Construction ◽

Temporal Changes ◽

Moisture Index ◽

Buffer Zones ◽

Wide Range ◽

Spatial And Temporal Changes

Highways provide key social and economic functions but generate a wide range of environmental consequences that are poorly quantified and understood. Here, we developed a before–during–after control-impact remote sensing (BDACI-RS) approach to quantify the spatial and temporal changes of environmental impacts during and after the construction of the Wujing Highway in China using three buffer zones (0–100 m, 100–500 m, and 500–1000 m). Results showed that land cover composition experienced large changes in the 0–100 m and 100–500 m buffers while that in the 500–1000 m buffer was relatively stable. Vegetation and moisture conditions, indicated by the normalized difference vegetation index (NDVI) and the normalized difference moisture index (NDMI), respectively, demonstrated obvious degradation–recovery trends in the 0–100 m and 100–500 m buffers, while land surface temperature (LST) experienced a progressive increase. The maximal relative changes as annual means of NDVI, NDMI, and LST were about −40%, −60%, and 12%, respectively, in the 0–100m buffer. Although the mean values of NDVI, NDMI, and LST in the 500–1000 m buffer remained relatively stable during the study period, their spatial variabilities increased significantly after highway construction. An integrated environment quality index (EQI) showed that the environmental impact of the highway manifested the most in its close proximity and faded away with distance. Our results showed that the effect distance of the highway was at least 1000 m, demonstrated from the spatial changes of the indicators (both mean and spatial variability). The approach proposed in this study can be readily applied to other regions to quantify the spatial and temporal changes of disturbances of highway systems and subsequent recovery.

Download Full-text

An LSWI-Based Method for Mapping Irrigated Areas in China Using Moderate-Resolution Satellite Data

Remote Sensing ◽

10.3390/rs12244181 ◽

2020 ◽

Vol 12 (24) ◽

pp. 4181

Author(s):

Kunlun Xiang ◽

Wenping Yuan ◽

Liwen Wang ◽

Yujiao Deng

Keyword(s):

Land Surface ◽

Vegetation Index ◽

Normalized Difference Vegetation Index ◽

Statistical Data ◽

Spatial Information ◽

Hydrological Cycle ◽

Mainland China ◽

Mean Annual Precipitation ◽

Irrigated Area ◽

Moderate Resolution

Accurate spatial information about irrigation is crucial to a variety of applications, such as water resources management, water exchange between the land surface and atmosphere, climate change, hydrological cycle, food security, and agricultural planning. Our study proposes a new method for extracting cropland irrigation information using statistical data, mean annual precipitation and Moderate Resolution Imaging Spectroradiometer (MODIS) land cover type data and surface reflectance data. The approach is based on comparing the land surface water index (LSWI) of cropland pixels to that of adjacent forest pixels with similar normalized difference vegetation index (NDVI). In our study, we validated the approach over mainland China with 612 reference samples (231 irrigated and 381 non-irrigated) and found the accuracy of 62.09%. Validation with statistical data also showed that our method explained 86.67 and 58.87% of the spatial variation in irrigated area at the provincial and prefecture levels, respectively. We further compared our new map to existing datasets of FAO/UF, IWMI, Zhu and statistical data, and found a good agreement with the irrigated area distribution from Zhu’s dataset. Results show that our method is an effective method apply to mapping irrigated regions and monitoring their yearly changes. Because the method does not depend on training samples, it can be easily repeated to other regions.

Download Full-text

Evaluating Downscaling Factors of Microwave Satellite Soil Moisture Based on Machine Learning Method

Remote Sensing ◽

10.3390/rs13010133 ◽

2021 ◽

Vol 13 (1) ◽

pp. 133

Author(s):

Hao Sun ◽

Yajing Cui

Keyword(s):

Machine Learning ◽

Soil Moisture ◽

Land Surface ◽

Regional Scale ◽

Support Vector ◽

Machine Learning Method ◽

Learning Method ◽

Feed Forward Neural Network ◽

Comparative Performance ◽

Geographical Factors

Downscaling microwave remotely sensed soil moisture (SM) is an effective way to obtain spatial continuous SM with fine resolution for hydrological and agricultural applications on a regional scale. Downscaling factors and functions are two basic components of SM downscaling where the former is particularly important in the era of big data. Based on machine learning method, this study evaluated Land Surface Temperature (LST), Land surface Evaporative Efficiency (LEE), and geographical factors from Moderate Resolution Imaging Spectroradiometer (MODIS) products for downscaling SMAP (Soil Moisture Active and Passive) SM products. This study spans from 2015 to the end of 2018 and locates in the central United States. Original SMAP SM and in-situ SM at sparse networks and core validation sites were used as reference. Experiment results indicated that (1) LEE presented comparative performance with LST as downscaling factors; (2) adding geographical factors can significantly improve the performance of SM downscaling; (3) integrating LST, LEE, and geographical factors got the best performance; (4) using Z-score normalization or hyperbolic-tangent normalization methods did not change the above conclusions, neither did using support vector regression nor feed forward neural network methods. This study demonstrates the possibility of LEE as an alternative of LST for downscaling SM when there is no available LST due to cloud contamination. It also provides experimental evidence for adding geographical factors in the downscaling process.

Download Full-text

A Semi-Automated Object-Based Gully Networks Detection Using Different Machine Learning Models: A Case Study of Bowen Catchment, Queensland, Australia

Sensors ◽

10.3390/s19224893 ◽

2019 ◽

Vol 19 (22) ◽

pp. 4893 ◽

Cited By ~ 22

Author(s):

Hejar Shahabi ◽

Ben Jarihani ◽

Sepideh Tavakkoli Piralilou ◽

David Chittleborough ◽

Mohammadtaghi Avand ◽

...

Keyword(s):

Machine Learning ◽

Image Segmentation ◽

Vegetation Index ◽

Scale Parameter ◽

Slope Aspect ◽

Support Vector ◽

Topographic Wetness Index ◽

Slope Length ◽

Optimal Scale ◽

Object Based

Gully erosion is a dominant source of sediment and particulates to the Great Barrier Reef (GBR) World Heritage area. We selected the Bowen catchment, a tributary of the Burdekin Basin, as our area of study; the region is associated with a high density of gully networks. We aimed to use a semi-automated object-based gully networks detection process using a combination of multi-source and multi-scale remote sensing and ground-based data. An advanced approach was employed by integrating geographic object-based image analysis (GEOBIA) with current machine learning (ML) models. These included artificial neural networks (ANN), support vector machines (SVM), and random forests (RF), and an ensemble ML model of stacking to deal with the spatial scaling problem in gully networks detection. Spectral indices such as the normalized difference vegetation index (NDVI) and topographic conditioning factors, such as elevation, slope, aspect, topographic wetness index (TWI), slope length (SL), and curvature, were generated from Sentinel 2A images and the ALOS 12-m digital elevation model (DEM), respectively. For image segmentation, the ESP2 tool was used to obtain three optimal scale factors. On using object pureness index (OPI), object matching index (OMI), and object fitness index (OFI), the accuracy of each scale in image segmentation was evaluated. The scale parameter of 45 with OFI of 0.94, which is a combination of OPI and OMI indices, proved to be the optimal scale parameter for image segmentation. Furthermore, segmented objects based on scale 45 were overlaid with 70% and 30% of a prepared gully inventory map to select the ML models’ training and testing objects, respectively. The quantitative accuracy assessment methods of Precision, Recall, and an F1 measure were used to evaluate the model’s performance. Integration of GEOBIA with the stacking model using a scale of 45 resulted in the highest accuracy in detection of gully networks with an F1 measure value of 0.89. Here, we conclude that the adoption of optimal scale object definition in the GEOBIA and application of the ensemble stacking of ML models resulted in higher accuracy in the detection of gully networks.

Download Full-text

Temporal Variations and Associated Remotely Sensed Environmental Variables of Dengue Fever in Chitwan District, Nepal

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi7070275 ◽

2018 ◽

Vol 7 (7) ◽

pp. 275 ◽

Cited By ~ 1

Author(s):

Bipin Acharya ◽

Chunxiang Cao ◽

Min Xu ◽

Laxman Khanal ◽

Shahid Naeem ◽

...

Keyword(s):

Surface Temperature ◽

Dengue Fever ◽

Land Surface Temperature ◽

Land Surface ◽

Environmental Variables ◽

Vegetation Index ◽

Generalized Additive Model ◽

Additive Model ◽

Remotely Sensed ◽

Cross Correlation Analysis

Dengue fever is one of the leading public health problems of tropical and subtropical countries across the world. Transmission dynamics of dengue fever is largely affected by meteorological and environmental factors, and its temporal pattern generally peaks in hot-wet periods of the year. Despite this continuously growing problem, the temporal dynamics of dengue fever and associated potential environmental risk factors are not documented in Nepal. The aim of this study was to fill this research gap by utilizing epidemiological and earth observation data in Chitwan district, one of the frequent dengue outbreak areas of Nepal. We used laboratory confirmed monthly dengue cases as a dependent variable and a set of remotely sensed meteorological and environmental variables as explanatory factors to describe their temporal relationship. Descriptive statistics, cross correlation analysis, and the Poisson generalized additive model were used for this purpose. Results revealed that dengue fever is significantly associated with satellite estimated precipitation, normalized difference vegetation index (NDVI), and enhanced vegetation index (EVI) synchronously and with different lag periods. However, the associations were weak and insignificant with immediate daytime land surface temperature (dLST) and nighttime land surface temperature (nLST), but were significant after 4–5 months. Conclusively, the selected Poisson generalized additive model based on the precipitation, dLST, and NDVI explained the largest variation in monthly distribution of dengue fever with minimum Akaike’s Information Criterion (AIC) and maximum R-squared. The best fit model further significantly improved after including delayed effects in the model. The predicted cases were reasonably accurate based on the comparison of 10-fold cross validation and observed cases. The lagged association found in this study could be useful for the development of remote sensing-based early warning forecasts of dengue fever.

Download Full-text

Mapping of the Canopy Openings in Mixed Beech–Fir Forest at Sentinel-2 Subpixel Level Using UAV and Machine Learning Approach

Remote Sensing ◽

10.3390/rs12233925 ◽

2020 ◽

Vol 12 (23) ◽

pp. 3925

Author(s):

Ivan Pilaš ◽

Mateo Gašparović ◽

Alan Novkinić ◽

Damir Klobučar

Keyword(s):

Machine Learning ◽

Forest Canopy ◽

Vegetation Index ◽

Predictive Performance ◽

Spatial Extent ◽

Gradient Boosting ◽

Support Vector ◽

Stochastic Gradient Boosting ◽

Extreme Gradient Boosting ◽

Sentinel 2

The presented study demonstrates a bi-sensor approach suitable for rapid and precise up-to-date mapping of forest canopy gaps for the larger spatial extent. The approach makes use of Unmanned Aerial Vehicle (UAV) red, green and blue (RGB) images on smaller areas for highly precise forest canopy mask creation. Sentinel-2 was used as a scaling platform for transferring information from the UAV to a wider spatial extent. Various approaches to an improvement in the predictive performance were examined: (I) the highest R2 of the single satellite index was 0.57, (II) the highest R2 using multiple features obtained from the single-date, S-2 image was 0.624, and (III) the highest R2 on the multitemporal set of S-2 images was 0.697. Satellite indices such as Atmospherically Resistant Vegetation Index (ARVI), Infrared Percentage Vegetation Index (IPVI), Normalized Difference Index (NDI45), Pigment-Specific Simple Ratio Index (PSSRa), Modified Chlorophyll Absorption Ratio Index (MCARI), Color Index (CI), Redness Index (RI), and Normalized Difference Turbidity Index (NDTI) were the dominant predictors in most of the Machine Learning (ML) algorithms. The more complex ML algorithms such as the Support Vector Machines (SVM), Random Forest (RF), Stochastic Gradient Boosting (GBM), Extreme Gradient Boosting (XGBoost), and Catboost that provided the best performance on the training set exhibited weaker generalization capabilities. Therefore, a simpler and more robust Elastic Net (ENET) algorithm was chosen for the final map creation.

Download Full-text

SVM-Based Geospatial Prediction of Soil Erosion Under Static and Dynamic Conditioning Factors

MATEC Web of Conferences ◽

10.1051/matecconf/201820304004 ◽

2018 ◽

Vol 203 ◽

pp. 04004

Author(s):

Muhammad Raza Ul Mustafa ◽

Abdulkadir Taofeeq Sholagberu ◽

Khamaruzaman Wan Yusof ◽

Ahmad Mustafa Hashim ◽

Muhammad Waris Ali Khan ◽

...

Keyword(s):

Soil Erosion ◽

Watershed Management ◽

Land Use Planning ◽

Land Surface ◽

Vegetation Index ◽

Support Vector ◽

Susceptibility Map ◽

Rainfall Erosivity ◽

Kappa Index ◽

Conditioning Factors

Land degradation caused by soil erosion remains an important global issue due to its adverse consequences on food security and environment. Geospatial prediction of erosion through susceptibility analysis is very crucial to sustainable watershed management. Previous susceptibility studies devoid of some crucial conditioning factors (CFs) termed dynamic CFs whose impacts on the accuracy have not been investigated. Thus, this study evaluates erosion susceptibility under the influence of both non-redundant static and dynamic CFs using support vector machine (SVM), remote sensing and GIS. The CFs considered include drainage density, lineament density, length-slope and soil erodibility as non-redundant static factors, and land surface temperature, soil moisture index, vegetation index and rainfall erosivity as the dynamic factors. The study implements four kernel tricks of SVM with sequential minimal optimization algorithm as a classifier for soil erosion susceptibility modeling. Using area under the curve (AUC) and Cohen’s kappa index (k) as the validation criteria, the results showed that polynomial function had the highest performance followed by linear and radial basis function. However, sigmoid SVM underperformed having the lowest AUC and k values coupled with higher classification errors. The CFs’ weights were implemented for the development of soil erosion susceptibility map. The map would assist planners and decision makers in optimal land-use planning, prevention of soil erosion and its related hazards leading to sustainable watershed management.

Download Full-text