Using supervised learning to develop BaRAD, a 40-year monthly bias-adjusted global gridded radiation dataset

AbstractDiffuse solar radiation is an important, but understudied, component of the Earth’s surface radiation budget, with most global climate models not archiving this variable and a dearth of ground-based observations. Here, we describe the development of a global 40-year (1980–2019) monthly database of total shortwave radiation, including its diffuse and direct beam components, called BaRAD (Bias-adjusted RADiation dataset). The dataset is based on a random forest algorithm trained using Global Energy Balance Archive (GEBA) observations and applied to the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2) dataset at the native MERRA-2 resolution (0.5° by 0.625°). The dataset preserves seasonal, latitudinal, and long-term trends in the MERRA-2 data, but with reduced biases than MERRA-2. The mean bias error is close to 0 (root mean square error = 10.1 W m−2) for diffuse radiation and −0.2 W m−2 (root mean square error = 19.2 W m−2) for the total incoming shortwave radiation at the surface. Studies on atmosphere-biosphere interactions, especially those on the diffuse radiation fertilization effect, can benefit from this dataset.

Download Full-text

Solar radiation forecasting by using deep neural networks in Eski̇şehi̇r

Sigma Journal of Engineering and Natural Sciences – Sigma Mühendislik ve Fen Bilimleri Dergisi ◽

10.14744/sigma.2021.00005 ◽

2021 ◽

pp. 159-169

Author(s):

Mohammed QASEM

Keyword(s):

Neural Network ◽

Solar Radiation ◽

Solar Energy ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Clean Energy ◽

Mean Bias Error ◽

Bias Error ◽

Mean Square

According to the World Economic Outlook (WEO), the global demand for energy is presum- ably going to be increased due to growing the world’s population up during the upcoming two decades. As a result of that, apprehensions about environmental effects, which appear as a re- sult of greenhouse gases are grown and cleaner energy technologies are developed. This clearly shows that extended growth of the worldwide market share of clean energy. Solar energy is considered as one of the fundamental types of renewable energy. For this reason, the need for a predictive model that effectively observes solar energy conversion with high performance becomes urgent. In this paper, classic empirical, artificial neural network (ANN), deep neural network (DNN), and time series models are applied, and their results are compared to each other to find the most accurate model for daily global solar radiation (DGSR) estimation. In addition, four regression models have been developed and applied for DGSR estimation. The obtained results are evaluated and compared by the root mean square error (RMSE), rela- tive root mean square error (rRMSE), mean absolute error (MAE), mean bias error (MBE), t-statistic, and coefficient of determination (R2). Finally, simulation results provided that the best result is found by the DNN model.

Download Full-text

A High-Resolution Topographic Correction Method for Clear-Sky Solar Irradiance Derived with a Numerical Weather Prediction Model

Journal of Applied Meteorology and Climatology ◽

10.1175/2011jamc2571.1 ◽

2011 ◽

Vol 50 (12) ◽

pp. 2460-2472 ◽

Cited By ~ 30

Author(s):

José A. Ruiz-Arias ◽

David Pozo-Vázquez ◽

Vicente Lara-Fanego ◽

Francisco J. Santos-Alamillos ◽

J. Tovar-Pescador

Keyword(s):

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Numerical Weather Prediction ◽

Solar Irradiance ◽

Weather Prediction ◽

Mean Bias Error ◽

Bias Error ◽

Mean Square ◽

Numerical Weather

AbstractRugged terrain is a source of variability in the incoming solar radiation field, but the influence of terrain is still not properly included by most current numerical weather prediction (NWP) models. In this work, a downscaling postprocessing method for NWP-model solar irradiance through terrain effects is presented. It allows one to decrease the estimation bias caused by terrain shading and sky-view reduction, and to account for elevation variability, surface orientation, and surface albedo. The method has been applied to a case study in southern Spain using the Weather Research and Forecasting (WRF) mesoscale model with a spatial resolution of 30 arc s, resulting in disaggregated maps of 3 arc s. The validation was based on a radiometric network made of eight stations located in the Natural Park of Sierra Mágina over an area of roughly 30 × 35 km2 and 12 carefully selected cloudless days during a year. Three of the stations were equipped with tilted pyranometers. Their inclination and aspect were visually adjusted to the inclination and aspect of the local terrain and then carefully measured. For horizontal surface, the downscaled irradiance has proven to reduce the root-mean-square error of the WRF model by 20% to about 25 W m−2 in winter and autumn and 60 W m−2 in spring and summer. For tilted surface, downscaling to different spatial resolutions resulted in the best performance for 9 arc s, with root-mean-square error of 45% (57 W m−2) and a mean bias error close to zero.

Download Full-text

Performance Analysis of Clear Sky Global Horizontal Irradiance Models: Simple Models Adapted for Local Conditions

Journal of Renewable Energy ◽

10.1155/2021/4369959 ◽

2021 ◽

Vol 2021 ◽

pp. 1-12

Author(s):

Nicholas Kwarikunda ◽

Zivayi Chiguvare

Keyword(s):

Solar Energy ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Measured Data ◽

Coefficient Of Determination ◽

Mean Bias Error ◽

Bias Error ◽

Mean Square ◽

Clear Sky

Evaluation of the maximum solar energy potential of a given area for possible deployment of solar energy technologies requires assessment of clear sky solar irradiance for the region under consideration. Such localized assessment is critical for optimal sizing of the technology to be deployed in order to realize the anticipated output. As the measurements are not always available where they are needed, models may be used to estimate them. In this study, three different models were adapted for the geographical location of the area under study and used to estimate clear sky global horizontal irradiance (GHI) at three locations in the subtropical desert climate of Namibia. The three models, selected on the basis of input requirements, were used to compute clear sky GHI at Kokerboom, Arandis, and Auas. The models were validated and evaluated for performance using irradiance data measured at each of the sites for a period of three years by computing statistical parameters such as mean bias error (MBE), root mean square error (RMSE), and the coefficient of determination (R2), normalized MBE, and normalized RMSE. Comparative results between modelled and measured data showed that the models fit well the measured data, with normalized root mean square error values in the range 4–8%, while the R2 value was above 98% for the three models. The adapted models can thus be used to compute clear sky GHI at these study areas as well as in other regions with similar climatic conditions.

Download Full-text

REDES NEURAIS ARTIFICIAIS, REGRESSÃO E MÉTODOS EMPÍRICOS PARA A MODELAGEM DA EVAPOTRANSPIRAÇÃO DE REFERÊNCIA NA CIDADE DE INHAMBANE, MOÇAMBIQUE

Irriga ◽

10.15809/irriga.2019v24n4p802-816 ◽

2019 ◽

Vol 24 (4) ◽

pp. 802-816

Author(s):

BARTOLOMEU FÉLIX TANGUNE ◽

Rodrigo Máximo Sánchez Román

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

Root Mean Square Error ◽

Mean Square Error ◽

Reference Evapotranspiration ◽

Mean Bias Error ◽

Bias Error ◽

Mean Square ◽

Empirical Methods ◽

E Mail

REDES NEURAIS ARTIFICIAIS, REGRESSÃO E MÉTODOS EMPÍRICOS PARA A MODELAGEM DA EVAPOTRANSPIRAÇÃO DE REFERÊNCIA NA CIDADE DE INHAMBANE, MOÇAMBIQUE BARTOLOMEU FÉLIX TANGUNE1 E RODRIGO MÁXIMO SÁNCHEZ ROMÁN2 1 Departamento de Engenharia Rural, Escola Superior de Desenvolvimento Rural, Universidade Eduardo Mondlane, Vilankulo, Inhambane, Moçambique. E-mail: [email protected]. 2 Departamento de Engenharia Rural, Faculdade de Ciências Agronômicas, Universidade Estadual Paulista (UNESP) Campus de Botucatu. Fazenda Experimental Lageado, Avenida Universitária, nº 3780, Altos do Paraíso, CEP: 18610-034, Botucatu – SP. Brasil. E-mail: [email protected] 1 RESUMO Estimativa precisa da evapotranspiração de referência (ETo) é importante para dimensionar e fazer manejo de sistemas de irrigação. Métodos de estimativa da ETo (11 métodos empíricos, 10 modelos de regressão múltipla: RLM e 10 redes neurais artificias: RNAs) foram avaliados em relação ao método padrão de Penman Monteith FAO 56, utilizando os seguintes índices: MBE (Mean Bias Error), RMSE (Root Mean Square Error) e R2, sendo RMSE utilizado como critério principal de seleção dos métodos. A significância dos métodos foi avaliada com base no teste t utilizando dados de 1985 a 2009. Os dados meteorológicos utilizados (temperatura máxima: Tmax, temperatura mínima: Tmin e temperatura média: T, umidade relativa, velocidade do vento e insolação) são da estação meteorológica convencional da cidade de Inhambane, Moçambique. Os resultados mostraram que o modelo RLM4 apresentou melhor desempenho (MBE = 0,01 mm.d-1; RMSE = 0,15 mm.d-1; R2 = 0,99). Na falta da radiação solar global, os modelos RLM6 (MBE = -0,01 mm.d-1; RMSE = 0,23 mm.d-1; R2 = 0,97) e RLM10 (MBE = 0,01 mm.d-1; RMSE = 0,23 mm.d-1; R2 = 0,97) podem ser utilizados e exigem a medição da T, Tmax e Tmin, respectivamente. Esses modelos não foram estatisticamente diferentes do método padrão. Palavras-chave: evapotranspiração, regressão múltipla, redes neurais. TANGUNE, B. F.; SÁNCHEZ-ROMÁN, R. M. ARTIFICIAL NEURAL NETWORKS, REGRESSION AND EMPIRICAL METHODS FOR REFERENCE EVAPOTRANSPIRATION MODELING IN INHAMBANE CITY, MOZAMBIQUE 2 ABSTRACT Precise estimation of reference evapotranspiration (ETo) is important for designing and managing irrigation systems. Methods of ETo estimation (11 empirical methods, 10 multiple regression models: RLM and 10 artificial neural networks: RNAs) were evaluated against Penman Monteith FAO 56 method using the following indexes: MBE (Mean Bias Error), RMSE (Root Mean Square Error) and R2, and RMSE was used as the main criterion of method selection. The significance of the methods was evaluated on the basis of the t test using data from 1985 to 2009. The meteorological data used (maximum temperature: Tmax, minimum temperature: Tmin and average temperature: T, relative air humidity, wind speed and solar brightness), from 1985 to 2009, are from the conventional meteorological station of the city of Inhambane, Mozambique. The results showed that the RLM4 model presented better performance (MBE = 0.01 mm.d-1; RMSE = 0.15 mm.d-1; R2 = 0.99). In the absence of global solar radiation, RLM6 (MBE = -0.01 mm.d-1; RMSE = 0.23 mm.d-1; R2 = 0.97) and RLM10 (MBE = 0.01 mm. d-1; RMSE = 0.23 mm.d-1; R2 = 0.97) can be used, which require measurement of T, and Tmax and Tmin, respectively. These models were not statistically different from the standard method. Keywords: evapotranspiration, multiple regression, neural networks.

Download Full-text

Long-Term Global Solar Radiation Prediction in 25 Cities in Morocco Using the FFNN-BP Method

Frontiers in Energy Research ◽

10.3389/fenrg.2021.733842 ◽

2021 ◽

Vol 9 ◽

Author(s):

Brahim Belmahdi ◽

Mohamed Louzazni ◽

Mohamed Akour ◽

Daniel Tudor Cotfas ◽

Petru Adrian Cotfas ◽

...

Keyword(s):

Solar Radiation ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Global Solar Radiation ◽

Mean Bias Error ◽

Mean Square ◽

Solar Declination ◽

Input Parameters ◽

Statistical Metrics

This article presents different combinations of input parameters based on an intelligent technique, using neural networks to predict daily global solar radiation (GSR) for twenty-five Moroccan cities. The collected measured data are available for 365 days and 25 stations around Morocco. Different input parameters are used, such as clearness index KT, day number, the length of the day, minimal temperature Tmin, maximal temperature Tmax, average temperature Taverage, difference temperature ΔT, ratio temperature T-Ratio, average relative humidity RH, solar radiation at the top outside atmosphere TOA, average wind speed Ws, altitude, longitude, latitude, and solar declination. A different combination was employed to predict daily GSR for the considered locations in order to find the most adequate input parameter that can be used in the prediction procedure. Several statistical metrics are applied to evaluate the performance of the obtained results, such as coefficients of determination (R2), mean absolute percentage error (MAPE), root mean square error (RMSE), normalized root mean square error (NRMSE), mean bias error (MBE), test statistic (TS), linear regression coefficients (the slope “a” and the constant “b”), and standard deviation (σ). It is found that the usage of input parameters gives highly accurate results in the artificial neural network (FFNN-BP) model, obtaining the lowest value of the statistical metrics. The results showed the best input of 25 locations, 12 inputs for Er-Rachidia, Marrakech, Medilt, Taza, Oujda, Nador, Tetouan, Tanger, Al-Auin, Dakhla, Settat, and Safi, seven inputs for Fes, Ifrane, Beni-Mellal, and Meknes, six inputs for Agadir and Rabat, five inputs for Sidi Ifni, Essaouira, Casablanca and Kenitra, four inputs for Ouarzazate, Lareche, and Al-Hoceima. In terms of accuracy, R2 of the selected best inputs parameters varies between 0.9860% and 0.9920%, the range value of MBE (%) being from −0.1076% to −0.5931%, the RMSE between 0.1990 and 0.4580%, the range value of the NRMSE between 0.0355 and 0.8938, and the lowest value MAPE between 0.0019 and 0.0060%. This technique could be used to predict other parameters for locations where measurement instrumentation is unavailable or costly to obtain.

Download Full-text

Validation of the new HelioClim-3 version 4 real-time and short-term forecast service using 14 BSRN stations

Advances in Science and Research ◽

10.5194/asr-13-129-2016 ◽

2016 ◽

Vol 13 ◽

pp. 129-136 ◽

Cited By ~ 7

Author(s):

Claire Thomas ◽

Laurent Saboret ◽

Etienne Wey ◽

Philippe Blanc ◽

Lucien Wald

Keyword(s):

Root Mean Square Error ◽

Real Time ◽

Mean Square Error ◽

Correlation Coefficient ◽

Root Mean Square ◽

Relative Bias ◽

Surface Radiation ◽

Solar Irradiation ◽

Mean Square ◽

Temporal Horizon

Abstract. Meteosat Second Generation (MSG) satellite images acquired every 15 min during daytime are currently processed by the Heliosat-2 method every night to generate the HelioClim-3 (HC3) database of the surface solar irradiation for the day before. A new service is proposed based on version 4 of HC3 (HC3v4) that offers real-time and forecasted irradiation for horizons up to a few hours. The service is based on a local persistence of the clear-sky index. Its results were compared to coincident high quality 15 min global irradiations measured in fourteen stations belonging to the Baseline Surface Radiation Network (BSRN). For forecasts for a temporal horizon of 15 min ahead, the relative bias and root mean square error (RMSE) range respectively from 0 to 2 %, and 20 to 23 % for most stations. The correlation coefficient ranges from 0.94 to 0.95. These performances are similar to HC3v4 for the same stations. Expectedly, the quality of the forecasts degrades as the temporal horizon increases. For 1 h ahead forecasts of 15 min irradiation, the relative bias, root mean square error (RMSE) and correlation coefficient range respectively from −3 to 1 %, 30 to 37 %, and 0.90 to 0.91.

Download Full-text

Estimation of Apple Flowering Frost Loss for Fruit Yield Based on Gridded Meteorological and Remote Sensing Data in Luochuan, Shaanxi Province, China

Remote Sensing ◽

10.3390/rs13091630 ◽

2021 ◽

Vol 13 (9) ◽

pp. 1630

Author(s):

Yaohui Zhu ◽

Guijun Yang ◽

Hao Yang ◽

Fa Zhao ◽

Shaoyu Han ◽

...

Keyword(s):

Remote Sensing ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Minimum Temperature ◽

Price Regulation ◽

Flowering Period ◽

Mean Square ◽

Prevention Measures ◽

Agricultural Insurance

With the increase in the frequency of extreme weather events in recent years, apple growing areas in the Loess Plateau frequently encounter frost during flowering. Accurately assessing the frost loss in orchards during the flowering period is of great significance for optimizing disaster prevention measures, market apple price regulation, agricultural insurance, and government subsidy programs. The previous research on orchard frost disasters is mainly focused on early risk warning. Therefore, to effectively quantify orchard frost loss, this paper proposes a frost loss assessment model constructed using meteorological and remote sensing information and applies this model to the regional-scale assessment of orchard fruit loss after frost. As an example, this article examines a frost event that occurred during the apple flowering period in Luochuan County, Northwestern China, on 17 April 2020. A multivariable linear regression (MLR) model was constructed based on the orchard planting years, the number of flowering days, and the chill accumulation before frost, as well as the minimum temperature and daily temperature difference on the day of frost. Then, the model simulation accuracy was verified using the leave-one-out cross-validation (LOOCV) method, and the coefficient of determination (R2), the root mean square error (RMSE), and the normalized root mean square error (NRMSE) were 0.69, 18.76%, and 18.76%, respectively. Additionally, the extended Fourier amplitude sensitivity test (EFAST) method was used for the sensitivity analysis of the model parameters. The results show that the simulated apple orchard fruit number reduction ratio is highly sensitive to the minimum temperature on the day of frost, and the chill accumulation and planting years before the frost, with sensitivity values of ≥0.74, ≥0.25, and ≥0.15, respectively. This research can not only assist governments in optimizing traditional orchard frost prevention measures and market price regulation but can also provide a reference for agricultural insurance companies to formulate plans for compensation after frost.

Download Full-text

Low Cost Automatic Reconstruction of Tree Structure by AdQSM with Terrestrial Close-Range Photogrammetry

Forests ◽

10.3390/f12081020 ◽

2021 ◽

Vol 12 (8) ◽

pp. 1020

Author(s):

Yanqi Dong ◽

Guangpeng Fan ◽

Zhiwu Zhou ◽

Jincheng Liu ◽

Yongguo Wang ◽

...

Keyword(s):

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Low Cost ◽

Reference Value ◽

Tree Height ◽

Point Clouds ◽

Tree Structure ◽

Mean Square ◽

Tree Model

The quantitative structure model (QSM) contains the branch geometry and attributes of the tree. AdQSM is a new, accurate, and detailed tree QSM. In this paper, an automatic modeling method based on AdQSM is developed, and a low-cost technical scheme of tree structure modeling is provided, so that AdQSM can be freely used by more people. First, we used two digital cameras to collect two-dimensional (2D) photos of trees and generated three-dimensional (3D) point clouds of plot and segmented individual tree from the plot point clouds. Then a new QSM-AdQSM was used to construct tree model from point clouds of 44 trees. Finally, to verify the effectiveness of our method, the diameter at breast height (DBH), tree height, and trunk volume were derived from the reconstructed tree model. These parameters extracted from AdQSM were compared with the reference values from forest inventory. For the DBH, the relative bias (rBias), root mean square error (RMSE), and coefficient of variation of root mean square error (rRMSE) were 4.26%, 1.93 cm, and 6.60%. For the tree height, the rBias, RMSE, and rRMSE were—10.86%, 1.67 m, and 12.34%. The determination coefficient (R2) of DBH and tree height estimated by AdQSM and the reference value were 0.94 and 0.86. We used the trunk volume calculated by the allometric equation as a reference value to test the accuracy of AdQSM. The trunk volume was estimated based on AdQSM, and its bias was 0.07066 m3, rBias was 18.73%, RMSE was 0.12369 m3, rRMSE was 32.78%. To better evaluate the accuracy of QSM’s reconstruction of the trunk volume, we compared AdQSM and TreeQSM in the same dataset. The bias of the trunk volume estimated based on TreeQSM was −0.05071 m3, and the rBias was −13.44%, RMSE was 0.13267 m3, rRMSE was 35.16%. At 95% confidence interval level, the concordance correlation coefficient (CCC = 0.77) of the agreement between the estimated tree trunk volume of AdQSM and the reference value was greater than that of TreeQSM (CCC = 0.60). The significance of this research is as follows: (1) The automatic modeling method based on AdQSM is developed, which expands the application scope of AdQSM; (2) provide low-cost photogrammetric point cloud as the input data of AdQSM; (3) explore the potential of AdQSM to reconstruct forest terrestrial photogrammetric point clouds.

Download Full-text