Related Experiment Video
Updated: Apr 28, 2026
09:55
Surface Renewal: An Advanced Micrometeorological Method for Measuring and Processing Field-Scale Energy Flux Density Data
Published on: December 12, 2013
8.1K
Performance evaluation of machine learning algorithms for estimating reference evapotranspiration based on NASA POWER
Oluwaseun Temitope Faloye1, Grace Awotoye2,3, Oluwadamilare Oluwasegun Eludire4
1Department of Water Resources Management and Agrometeorology, Federal University, Oye-Ekiti, Ekiti, Nigeria.
Frontiers in Artificial Intelligence
|April 27, 2026
Summary
This study shows that fine Gaussian support vector machine (SVM) models using NASA POWER data accurately estimate reference evapotranspiration (ETo). This method is effective for water management where ground data is scarce.
Area of Science:
- Hydrology
- Agricultural Meteorology
- Machine Learning Applications
Background:
- Reference evapotranspiration (ETo) is crucial for water resource management.
- The Penman-Monteith (PM) method is the standard for ETo estimation but requires extensive data.
- There is a need for accurate ETo prediction methods in data-scarce regions.
Purpose of the Study:
- To evaluate machine learning (ML) models, specifically Support Vector Machines (SVM) and Decision Trees (DT) with ensembles, for predicting ETo.
- To utilize NASA NEX-GDDP data as input for ML models in ETo estimation.
- To compare the performance of various SVM kernels and DT ensembles in diverse Nigerian climatic conditions.
Main Methods:
- Trained ML models using average monthly climatic data (temperature, humidity, wind speed) from NASA POWER.
- Calculated ETo using ground-observed data via the PM method as the target variable.
- Employed SVM models (linear, quadratic, cubic, Gaussian variants) and DT models (fine, medium, coarse, bagged, boosted ensembles).
- Validated models using 30% training and 70% testing data splits across dry, wet, and moderate weather locations in Nigeria.
Main Results:
- The fine Gaussian SVM (FG SVM) model demonstrated superior performance in ETo estimation.
- FG SVM achieved a Root Mean Square Error (RMSE) of 0.38 mm (training) and 0.599 mm (testing).
- High coefficients of determination (r²) of 0.87 (training) and 0.72 (validation) were recorded for FG SVM.
- FG SVM outperformed all other tested models across different climatic zones.
Conclusions:
- The integration of NASA POWER data with the FG SVM model provides an accurate and robust method for estimating ETo.
- This ML approach is highly valuable for water resource management in regions with limited or unavailable ground-based climatic data.
- The study highlights the potential of advanced ML techniques in overcoming data limitations in hydrological and agricultural applications.
Related Concept Videos
01:29Application of Linearization and Approximation
192
A drone flying through complex terrain often relies on more than one sensing method to estimate small changes in altitude. Along with direct measurements, air pressure provides a useful indirect indicator of vertical movement. Atmospheric pressure decreases as altitude increases, and this relationship is commonly described using an exponential model. Although accurate, converting pressure measurements into altitude values requires calculations that are too complex to perform repeatedly during...
192
01:17Precipitation and Co-precipitation
4.8K
Precipitation and coprecipitation methods can be used to separate a mixture of ions in a solution. In qualitative inorganic analysis, ions that form sparingly soluble precipitates with the same reagent are separated based on the differences in solubility products. For example, consider the separation of Cu(II) and Fe(II) ions by precipitation as insoluble sulfides. First, copper(II) sulfide is precipitated by the addition of acidic H2S, where the dissociation of H2S is suppressed. Adding H2S...
4.8K