Shallow Groundwater Levels in a Wet Prairie: Using Machine Learning to Predict Water Level Changes in Northwest Ohio

More, P. R.; Gomezdelcampo, Enrique; Roberts, Sheila J.

doi:10.1130/abs/2018nc-312966

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our study, we observed instability in the FNN model, with no significant increase in RMSE from the training to testing stages (Table 2). It is expected for the NARX model's performance to vary during the training (open-loop) and testing (closed-loop) stages due to its dependence on different loop designs (More et al, 2018;Wunsch et al, 2018).…”

Section: Discussionmentioning

confidence: 99%

Artificial Neural Networks Estimate Evapotranspiration For Miscanthus × Giganteus As Effectively As Empirical Model But With Fewer Inputs

Sungur,

Moore,

Bernacchi

et al. 2024

Preprint

View full text Add to dashboard Cite

Estimating actual evapotranspiration (ET) is particularly crucial for addressing how vegetation affects the water balance of ecosystems. ET estimation can be complex with empirical models due to their many parameters and reliance on aridity. In contrast, artificial neural networks (ANNs) could potentially estimate ET with fewer and more common meteorological parameters. In this study, we trained two ANNs, one using a feed-forward approach (FFN) and the other a nonlinear auto-regressive network (NARX), to predict ET and compared them to the commonly used empirical model Granger and Gray (GG). We trained our models on a nine-year eddy covariance (EC) dataset for Miscanthus × giganteus (M. × giganteus) from Illinois (UIEF), then tested them using out-of-sample data from both UIEF and a different location in Iowa (SABR) to compare the accuracy of FFN, NARX, and GG models in estimating daily ET. A combination of air temperature (Ta) and solar radiation (Rs) was chosen as inputs due to the highest R2 for FFN (R2= 0.79, 0.81, and 0.79 for training, testing, and validation, respectively) and only Ta for NARX (R2= 0.70 for out-of-sample validation). The predictive power of the FFN model was superior to the NARX and GG models at the UIEF site (R2= 0.84, 0.70, and 0.83 for out-of-sample validation, respectively). Our analysis showed that ANN approaches are as accurate as empirical approaches for estimating ET but use fewer inputs.

show abstract

Section: Discussionmentioning

confidence: 99%

Artificial Neural Networks Estimate Evapotranspiration For Miscanthus × Giganteus As Effectively As Empirical Model But With Fewer Inputs

Sungur,

Moore,

Bernacchi

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…ANNs have been successfully applied to several agricultural engineering problems, including yield prediction such as corn yield (Uhrig et al, 1992; Uno et al, 2005), soybean growth (Zhang et al, 2009) and wheat yield (Ruβ et al, 2008). In water management, ANNs have been applied to simulate the groundwater levels in coastal aquifers (Taormina et al, 2012), to predict water table response to change in precipitation (More, 2018), to predict drainage water and groundwater salinity at various drain depths and spacing (Nozari & Azadi, 2017) and to estimate evapotranspiration (Feng et al, 2017). In soil management studies, ANNs have been used to determine soil temperature (Nahvi et al, 2016), to estimate total soil nitrogen, organic carbon and moisture content (Morellos et al, 2016) and to estimate soil hydraulic properties such as soil water content, saturated and unsaturated hydraulic conductivity (Ellafi et al, 2021; Schaap et al, 2001) and reduction in hydraulic conductivity (Ezlit et al, 2014).…”

Section: Methods To Obtain Input Data For Drainmod In Data‐poor Arid ...mentioning

confidence: 99%

Drainage models: An evaluation of their applicability for the design of drainage systems in arid regions

Ellafi

Simmons

Deeks

2023

Water & Environment J

View full text Add to dashboard Cite

Only 5%–10% of irrigated lands in least developed countries (LDCs) are currently drained. Although drainage simulation models (DSMs) are used to evaluate alternative designs, it is unclear which drainage model is suitable for LDCs' arid and semi‐arid regions. This study evaluates selected DSMs (ADAPT, RZWQM2, DRAINMOD, EPIC, HYDRUS‐1D, WaSim and SWAP) and critically assesses their applicability to arid and semi‐arid areas. Also, establish and apply selection criteria based on the availability of data in LDCs with Libya as a case study, and identify the most suitable model for application in Libya. DRAINMOD had the highest overall score, and alternative methods to predict missing input parameters for DRAINMOD are discussed. Evaluating the feasibility of using predicted input parameters for DSMs to design drainage systems in LDCs would help farmers, planners and decision‐makers to reduce the overall cost of drainage system and, also, make DRAINMOD a more accessible tool to evaluate different drainage designs.

show abstract

Shallow Groundwater Levels in a Wet Prairie: Using Machine Learning to Predict Water Level Changes in Northwest Ohio

Cited by 2 publications

References 0 publications

Artificial Neural Networks Estimate Evapotranspiration For Miscanthus × Giganteus As Effectively As Empirical Model But With Fewer Inputs

Artificial Neural Networks Estimate Evapotranspiration For Miscanthus × Giganteus As Effectively As Empirical Model But With Fewer Inputs

Drainage models: An evaluation of their applicability for the design of drainage systems in arid regions

Contact Info

Product

Resources

About