Article
Agronomy
Offer Rozenstein, Lior Fine, Nitzan Malachy, Antoine Richard, Cedric Pradalier, Josef Tanny
Summary: Advancements in remote sensing and machine learning have shown potential in improving irrigation use efficiency. This study calibrated and tested two methods to determine irrigation doses in tomato fields. Results demonstrate that both the UAV and ANN methods estimated evapotranspiration and irrigation doses with near-perfect agreement to best-practice irrigation.
AGRICULTURAL WATER MANAGEMENT
(2023)
Article
Agronomy
Juan Antonio Bellido-Jimenez, Javier Estevez, Amanda Penelope Garcia-Marin
Summary: The study evaluates various neural network approaches for estimating Reference Evapotranspiration (ET0) and finds that the use of new variables such as EnergyT and Hourmin can improve accuracy, especially in humid regions. Among the models tested, Multilayer Perceptron (MLP) and Extreme Learning Machine (ELM) stand out as reliable options for accurately estimating ET0 in locations with limited data acquisition.
AGRICULTURAL WATER MANAGEMENT
(2021)
Article
Engineering, Civil
Ammara Talib, Ankur R. Desai, Jingyi Huang, Tim J. Griffis, David E. Reed, Jiquan Chen
Summary: Evapotranspiration prediction and forecasting are crucial for improving water use efficiency in agriculture, and the use of random forest models has shown better performance in estimating daily ET compared to LSTM models. Vapor pressure and crop coefficients are important predictors for irrigated crops, while short wave radiation and enhanced vegetation index are key predictors for non-irrigated crops.
JOURNAL OF HYDROLOGY
(2021)
Article
Plant Sciences
Zewei Jiang, Shihong Yang, Shide Dong, Qingqing Pang, Pete Smith, Mohamed Abdalla, Jie Zhang, Guangmei Wang, Yi Xu
Summary: Cotton is widely used in various industries but faces threats from soil salinization. Drip irrigation plays a crucial role in improving water and fertilization efficiency. Accurate prediction of soil salinity and crop evapotranspiration is important for water management in arid and saline regions. We proposed a method based on machine learning to simulate soil salinity, evapotranspiration, and cotton yield using a global dataset.
FRONTIERS IN PLANT SCIENCE
(2023)
Article
Ecology
Antonino Pagano, Federico Amato, Matteo Ippolito, Dario De Caro, Daniele Croce, Antonio Motisi, Giuseppe Provenzano, Ilenia Tinnirello
Summary: Accurate estimation of actual evapotranspiration (ETa) is crucial for various environmental issues, and artificial intelligence-based models show promise as an alternative to traditional measurement techniques. This research evaluates two machine learning algorithms, Multi-Layer Perceptron (MLP) and Random Forest (RF), for predicting daily ETa in a citrus orchard. The best performance is achieved by the Random Forest method, with seven input features, obtaining a root mean square error (RMSE) of 0.39 mm/day and a coefficient of determination (R2) of 0.84. The results highlight the importance of using soil water content (SWC), weather, and satellite data together for improved evapotranspiration forecasts.
ECOLOGICAL INFORMATICS
(2023)
Article
Environmental Sciences
Yuan Liu, Qi Jiang, Qianyang Wang, Yongliang Jin, Qimeng Yue, Jingshan Yu, Yuexin Zheng, Weiwei Jiang, Xiaolei Yao
Summary: This study utilized remote sensing data to investigate the relationships among global extreme drought, evapotranspiration, and meteorological, hydrological, and botanical factors. Future global evapotranspiration was predicted and a method for deriving future evapotranspiration values was provided, showing significant value for climate change adaptation and drought warning.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Environmental Sciences
Josefina Mosre, Francisco Suarez
Summary: Estimating actual evapotranspiration (ETa) in arid regions is challenging due to its dynamic nature over time and space. The study found that available energy is the main meteorological variable controlling ETa, and incorporating remote sensing vegetation indices (VIs) improves ETa estimates. Regression equations performed best for monthly ETa estimates with the inclusion of VIs.
Article
Environmental Sciences
Xiaoman Jiang, Guoqiang Wang, Yuntao Wang, Jiping Yao, Baolin Xue, A. Yinglan
Summary: Evapotranspiration (ET) is crucial for understanding climate change, ecological problems, the water cycle, and hydrological processes. Machine learning algorithms have been used to estimate ET, but limitations in data resolution hinder the discovery of underlying patterns. In this study, a hybrid framework was developed to simulate ET in data-deficient areas using a coupled model and random forest. This approach successfully simulated monthly ET in the Inner Mongolia section of the Yellow River Basin from 1982 to 2020, with good results. The analysis also revealed temporal and spatial variations in ET and identified the important factors influencing ET in the region.
Article
Chemistry, Multidisciplinary
Jianwei Geng, Hengpeng Li, Wenfei Luan, Yunjie Shi, Jiaping Pang, Wangshou Zhang
Summary: This study investigated the potential of ensemble learning algorithms in predicting the daily evapotranspiration of tea plants. The results showed that the random forest model demonstrated superior performance compared to other algorithms.
APPLIED SCIENCES-BASEL
(2023)
Article
Environmental Sciences
Rafael Lopez-Olivari, Sigfredo Fuentes, Carlos Poblete-Echeverria, Valeria Quintulen-Ancapi, Leovijildo Medina
Summary: Evapotranspiration is an important component of agricultural water management systems, especially in water-limited environments. Evaluating different canopy resistance methods, it was found that the amphistomatous and hypostomatous approaches are the best options for estimating potato evapotranspiration. The ETcpLA method provided accurate estimates compared to the soil water balance method for different irrigation levels.
Article
Meteorology & Atmospheric Sciences
Xian Wang, Lei Zhong, Yaoming Ma, Yunfei Fu, Cunbo Han, Peizhen Li, Zixin Wang, Yuting Qi
Summary: This paper proposes an ETa product with hourly temporal resolution for the entire Tibetan Plateau (TP) using Fengyun-4A geostationary satellite data and a random forest (RF) model. Compared with other models, the RF model shows the best performance in estimating ETa. The total ETa for the entire TP is approximately 365.60 mm, and the total water amount evapotranspired from the TP surface is approximately 9811.01x10(8) t yr(-1). Additionally, the diurnal, monthly, and seasonal variations of ETa in different land cover types and climate zones over the TP are quantified.
ATMOSPHERIC RESEARCH
(2023)
Article
Plant Sciences
Rodrigo Filev Maia, Carlos Ballester Lurbe, John Hornbuckle
Summary: There is a growing interest in using the Internet of Things (IoT) in agriculture to acquire data for more efficient farm management. This study analyzed sensor data to identify the relationship between soil matric potential and crop evapotranspiration in cotton fields. Machine learning models were used to accurately estimate soil moisture from satellite data. The findings have promising applications in irrigation-decision systems.
FRONTIERS IN PLANT SCIENCE
(2022)
Article
Agronomy
Arti Kumari, Ashutosh Upadhyaya, Pawan Jeet, Nadhir Al-Ansari, Jitendra Rajput, Prem K. Sundaram, Kirti Saurabh, Ved Prakash, Anil K. Singh, Rohan K. Raman, Venkatesh Gaddikeri, Alban Kuriqi
Summary: In this study, actual crop evapotranspiration (ETa) and stagewise crop coefficient (K-c) of transplanted puddled rice were determined using a modified non-weighing paddy lysimeter. The results were compared with indirect methods, and it was found that the FAO Penman-Monteith equation performed well, while the pan evaporation approach underestimated the ETa. Furthermore, actual K-c values were obtained and compared with FAO values. This methodology can be used to improve irrigation scheduling under similar agro-climatic conditions.
Article
Agronomy
Ana Cristina Garcia-Vasquez, Esmaiil Mokari, Zohrab Samani, Alexander Fernald
Summary: The objective of this study is to calculate real-time pecan water use and improve irrigation efficiency using UAV thermal imaging when the canopy cover is variable. The proposed methodology successfully calculated ETa and on-farm irrigation efficiency in the area.
AGRICULTURAL WATER MANAGEMENT
(2022)
Article
Agronomy
Hemendra Kumar, Puneet Srivastava, Jasmeet Lamba, Bruno Lena, Efstathios Diamantopoulos, Brenda Ortiz, Bijoychandra Takhellambam, Guilherme Morata, Luca Bondesan
Summary: This study focused on determining zone-specific field capacity (FC) and irrigation thresholds using a negligible drainage flux (qfc) criterion in Alabama, USA. The results showed that the optimized FC values were more accurate than the raw values, and the proposed method can help improve irrigation management.
AGRICULTURAL WATER MANAGEMENT
(2023)
Article
Meteorology & Atmospheric Sciences
H. J. S. Fernando, I Gultepe, C. Dorman, E. Pardyjak, Q. Wang, S. W. Hoch, D. Richter, E. Creegan, S. Gabersek, T. Bullock, C. Hocut, R. Chang, D. Alappattu, R. Dimitrova, D. Flagg, A. Grachev, R. Krishnamurthy, D. K. Singh, I Lozovatsky, B. Nagare, A. Sharma, S. Wagh, C. Wainwright, M. Wroblewski, R. Yamaguchi, S. Bardoel, R. S. Coppersmith, N. Chisholm, E. Gonzalez, N. Gunawardena, O. Hyde, T. Morrison, A. Olson, A. Perelet, W. Perrie, S. Wang, B. Wauer
Summary: C-FOG is a comprehensive bi-national project focused on studying the formation, persistence, and dissipation of coastal fog. Through field observations and modeling, it integrates research across various processes, dynamics, microphysics, and thermodynamics to address the complexity of coastal fog. The project aims to identify and remedy numerical model deficiencies by utilizing a multiplatform framework for interpretation of field observations.
BULLETIN OF THE AMERICAN METEOROLOGICAL SOCIETY
(2021)
Article
Environmental Sciences
Florian Dupuy, Gert-Jan Duine, Pierre Durand, Thierry Hedde, Eric Pardyjak, Pierre Roubin
Summary: In regions with complex topography, it is challenging to forecast local flows accurately due to coarse resolution of operational models. A study utilized an artificial neural network (ANN) as a correcting tool to significantly improve forecast accuracy of low-level winds (both speed and direction) based on Weather Research and Forecasting (WRF) model simulations.
Article
Computer Science, Interdisciplinary Applications
Behnam Bozorgmehr, Pete Willemsen, Jeremy A. Gibbs, Rob Stoll, Jae-Jin Kim, Eric R. Pardyjak
Summary: QES-Winds is a numerical code designed for computing high-resolution wind fields in real time on large domains, utilizing variational analysis technique and GPU parallelization to accelerate the numerical solution, suitable for modeling a wide range of real-world problems.
ENVIRONMENTAL MODELLING & SOFTWARE
(2021)
Article
Meteorology & Atmospheric Sciences
Alexei O. Perelet, Ismail Gultepe, Sebastian W. Hoch, Eric R. Pardyjak
Summary: The study investigates the visibility and discrimination of fog and rain events using a two-wavelength scintillometer system. Results show that near-infrared and microwave radiation are attenuated to similar levels during fog and precipitation events.
BOUNDARY-LAYER METEOROLOGY
(2021)
Article
Meteorology & Atmospheric Sciences
Travis Morrison, Eric R. Pardyjak, Matthias Mauder, Marc Calaf
Summary: This study quantifies spatial heterogeneity using data from the Idealized Planar-Array experiment and analyzes the heat-flux imbalance. The results show different biases in the estimation of turbulence flux under different meteorological conditions, and indicate that mean air temperature heterogeneity leads to strong bulk advection and dispersive fluxes.
BOUNDARY-LAYER METEOROLOGY
(2022)
Article
Geosciences, Multidisciplinary
Chaoxun Hang, Holly J. Oldroyd, Marco G. Giometto, Eric R. Pardyjak, Marc B. Parlange
Summary: The study introduces a modified local-MOST stability-correction function, which can improve the modeling of katabatic flows by addressing the violation of traditional flux-gradient relations. By utilizing turbulence observations from slopes with different inclination angles, the proposed relation demonstrates better convergence and higher accuracy in data application, showing significant improvement compared to traditional parameterizations.
GEOPHYSICAL RESEARCH LETTERS
(2021)
Article
Environmental Sciences
Stephen Drake, Chad Higgins, Eric Pardyjak
Summary: The experiment shows that small-scale drainage features regulate the local cooling rate, while topographic slope and distance along the drainage influence the larger-scale cooling rate. The difference in cooling rate between local and basin-wide scales suggests that small-scale features have faster timescales that are most pronounced shortly after local sunset.
Article
Forestry
Matthew J. Moody, Jeremy A. Gibbs, Steven Krueger, Derek Mallia, Eric R. Pardyjak, Adam K. Kochanski, Brian N. Bailey, Rob Stoll
Summary: QES-Fire is a microscale wildfire model that dynamically couples the fire front to microscale winds, while being computationally efficient. The model represents complex fire fronts using a multiscale plume-merging methodology and shows good agreement with atmospheric large-eddy simulation and field experiment data.
INTERNATIONAL JOURNAL OF WILDLAND FIRE
(2022)
Article
Environmental Sciences
Nipun Gunawardena, Pierre Durand, Thierry Hedde, Florian Dupuy, Eric Pardyjak
Summary: This paper compares two computationally inexpensive methods for predicting and filling spatially varying meteorological variables in complex terrain. The methods, multivariable linear regression and artificial neural networks, are tested on real data to evaluate their performance.
Article
Environmental Sciences
Fabien Margairaz, Hanieh Eshagh, Arash Nemati Hayati, Eric R. Pardyjak, Rob Stoll
Summary: A new wake model for isolated trees is proposed and evaluated in this study. The model shows good accuracy in predicting wind speed and direction, with small deviations in wind speed around isolated trees. Furthermore, the impact of trees and tree wakes on the flow field is highly dependent on the proximity to buildings and building wakes in a complex urban environment.
Article
Meteorology & Atmospheric Sciences
Alexei O. Perelet, Helen C. Ward, Rob Stoll, Walter F. Mahaffee, Eric R. Pardyjak
Summary: Scintillometry is a non-invasive measurement technique used to acquire spatially-averaged surface heat and moisture fluxes. In this study, a two-wavelength scintillometry system was deployed to test an active vineyard, and the results showed that the vineyard was homogeneous at the spatial scales of the scintillometer path.
BOUNDARY-LAYER METEOROLOGY
(2022)
Article
Meteorology & Atmospheric Sciences
Marc Calaf, Nikki Vercauteren, Gabriel G. Katul, Marco G. Giometto, Travis J. Morrison, Fabien Margairaz, Vyacheslav Boyko, Eric R. Pardyjak
Summary: The time integration of the unsteady Reynolds-averaged Navier-Stokes equations is widely used in numerical weather prediction. This approach divides the flow into an ensemble mean and turbulence-related fluctuations, allowing closure schemes to describe their statistical properties. However, modelling challenges arise when unresolved fluctuations include non-turbulent structured motions, which can render conventional closure schemes ineffective. This study seeks to address these challenges by discussing theoretical tactics and considering the use of large-eddy simulations, direct numerical simulations, and field measurements.
BOUNDARY-LAYER METEOROLOGY
(2023)
Article
Meteorology & Atmospheric Sciences
Travis J. Morrison, Marc Calaf, Eric R. Pardyjak
Summary: The closure of the surface energy balance (SEB) has been a long-standing problem for the atmospheric boundary layer (ABL) community. This study compares the results of a three-dimensional SEB approach with the traditional one-dimensional SEB approach using data from the IPAQS 2019 field campaign. The results show that the three-dimensional approach improves the closure of the SEB during convective periods.
QUARTERLY JOURNAL OF THE ROYAL METEOROLOGICAL SOCIETY
(2023)
Article
Meteorology & Atmospheric Sciences
Dhiraj K. Singh, Spencer Donovan, Eric R. Pardyjak, Timothy J. Garrett
Summary: DEID is a novel instrument designed for measuring properties of precipitation, including mass, size, density, and type, as well as overall characteristics. It utilizes a thermal approach to measure the spatial dimensions of hydrometeors on a heated metal plate, offering a method for discriminating precipitation phase. The instrument provides accurate measurements of precipitation and shows good agreement with canonical results described in the literature.
ATMOSPHERIC MEASUREMENT TECHNIQUES
(2021)
Article
Geosciences, Multidisciplinary
Carlos Roman-Cascon, Marie Lothon, Fabienne Lohou, Oscar Hartogensis, Jordi Vila-Guerau de Arellano, David Pino, Carlos Yague, Eric R. Pardyjak
Summary: This study focuses on simulating surface heat fluxes in the interface between the Earth's surface and the atmosphere, and finds that using a more realistic surface representation, exploring mosaic approach, and adjusting vegetation type parameters can improve the model simulated fluxes and enhance the accuracy of the model in representing climate and weather changes.
GEOSCIENTIFIC MODEL DEVELOPMENT
(2021)