Automated Geospatial Models of Varying Complexities for Pine Forest Evapotranspiration Estimation with Advanced Data Mining

Abstract

The study goal was to develop automated user-friendly remote-sensing based evapotranspiration (ET) estimation tools: (i) artificial neural network (ANN) based models, (ii) ArcGIS-based automated geospatial model, and (iii) executable software to predict pine forest daily ET flux on a pixel- or plot average-scale. Study site has had long-term eddy-flux towers for ET measurements since 2006. Cloud-free Landsat images of 2006−2014 were processed using advanced data mining to obtain Principal Component bands to correlate with ET data. The regression model’s r2 was 0.58. The backpropagation neural network (BPNN) and radial basis function network (RBFN) models provided a testing/validation average absolute error of 0.18 and 0.15 Wm−2 and average accuracy of 81% and 85%, respectively. ANN models though robust, require special ANN software and skill to operate; therefore, automated geospatial model (toolbox) was developed on ArcGIS ModelBuilder as user-friendly alternative. ET flux map developed with model tool provided consistent ET patterns for landuses. The software was developed for lay-users for ET estimation.

A Prediction Model of Water In Situ Data Change under the Influence of Environmental Variables in Remote Sensing Validation

“…BPNN is one of the basic neural networks, it has excellent performance in dealing with multivariate and nonlinear relationships. Up to now, BPNN has been widely used in the retrieval of remote sensing parameters [38][39][40][41]. However, two shortcomings of BPNN restrict its application: (1) Convergence is slow during training, and (2) BPNN is too sensitive to the initial network weights, and different initial weights help the model converge to different local minimums [42].…”

Artificial intelligence and machine learning applications in forest management and biodiversity conservation

“…The study of ecohydrology, specifically forest hydrology, is utilizing advanced technologies like AI and ML to harness the full potential of data and gain novel insights concerning ecohydrological processes [62] . AI/ML techniques have been employed to calculate and simulate the amount of rainwater intercepted by forest canopies, the water content of canopies, the spatiotemporal patterns of soil moisture in vegetated regions, the evapotranspiration of land on a global and regional scale, the efficiency of water usage in terrestrial ecosystems, the storage of water in vegetation, the estimation of terrestrial and groundwater storage utilizing vegetation cover as an indicator, as well as the assessment of water stress in plants [63][64][65][66][67][68][69][70][71] . The proliferation of large hydrologic datasets obtained via remote sensing and data compilation has facilitated the integration of ML techniques into land surface modeling.…”

Potential for Artificial Intelligence (AI) and Machine Learning (ML) Applications in Biodiversity Conservation, Managing Forests, and Related Services in India

“…Tools such as AI and ML are being applied in the field of ecohydrology, including forest hydrology, to fully realize the potential of these data and obtain new insights into ecohydrological processes [80]. For instance, AI/ML methods have been used to estimate and model precipitation interception by forest canopies [81,82], canopy water content [83], spatiotemporal behavior of soil moisture in vegetated areas [84][85][86], global [87] and regional [88] terrestrial evapotranspiration, water-use efficiency in terrestrial ecosystems [89], vegetation water storage [90], terrestrial/groundwater storage [91] using vegetation cover as an indicator [92], and plant water stress [93]. The recent growth of big hydrologic data through remote sensing and data compilation has also fostered the adoption of ML in land surface modeling which simulates land surface processes including partitioning of water between land and atmosphere, such as in groundwater dynamics [94].…”