Estimation of hourly global photosynthetically active radiation using artificial neural network models

Extension of PAR Models under Local All-Sky Conditions to Different Climatic Zones

Four models for predicting Photosynthetically Active Radiation (PAR) were obtained through MultiLinear Regression (MLR) and an Artificial Neural Network (ANN) based on 10 meteorological indices previously selected from a feature selection algorithm. One model was developed for all sky conditions and the other three for clear, partial, and overcast skies, using a sky classification based on the clearness index (kt). The experimental data were recorded in Burgos (Spain) at ten-minute intervals over 23 months between 2019 and 2021. Fits above 0.97 and Root Mean Square Error (RMSE) values below 7.5% were observed. The models developed for clear and overcast sky conditions yielded better results. Application of the models to the seven experimental ground stations that constitute the Surface Radiation Budget Network (SURFRAD) located in different Köppen climatic zones of the USA yielded fitted values higher than 0.98 and RMSE values less than 11% in all cases regardless of the sky type.

Sub-Daily Natural CO2 Flux Simulation Based on Satellite Data: Diurnal and Seasonal Pattern Comparisons to Anthropogenic CO2 Emissions in the Greater Tokyo Area

During the last decade, advances in the remote sensing of greenhouse gas (GHG) concentrations by the Greenhouse Gases Observing SATellite-1 (GOSAT-1), GOSAT-2, and Orbiting Carbon Observatory-2 (OCO-2) have produced finer-resolution atmospheric carbon dioxide (CO2) datasets. These data are applicable for a top-down approach towards the verification of anthropogenic CO2 emissions from megacities and updating of the inventory. However, great uncertainties regarding natural CO2 flux estimates remain when back-casting CO2 emissions from concentration data, making accurate disaggregation of urban CO2 sources difficult. For this study, we used Moderate Resolution Imaging Spectroradiometer (MODIS) land products, meso-scale meteorological data, SoilGrids250 m soil profile data, and sub-daily soil moisture datasets to calculate hourly photosynthetic CO2 uptake and biogenic CO2 emissions with 500 m resolution for the Kantō Plain, Japan, at the center of which is the Tokyo metropolis. Our hourly integrated modeling results obtained for the period 2010–2018 suggest that, collectively, the vegetated land within the Greater Tokyo Area served as a daytime carbon sink year-round, where the hourly integrated net atmospheric CO2 removal was up to 14.15 ± 4.24% of hourly integrated anthropogenic emissions in winter and up to 55.42 ± 10.39% in summer. At night, plants and soil in the Greater Tokyo Area were natural carbon sources, with hourly integrated biogenic CO2 emissions equivalent to 2.27 ± 0.11%–4.97 ± 1.17% of the anthropogenic emissions in winter and 13.71 ± 2.44%–23.62 ± 3.13% in summer. Between January and July, the hourly integrated biogenic CO2 emissions of the Greater Tokyo Area increased sixfold, whereas the amplitude of the midday hourly integrated photosynthetic CO2 uptake was enhanced by nearly five times and could offset up to 79.04 ± 12.31% of the hourly integrated anthropogenic CO2 emissions in summer. The gridded hourly photosynthetic CO2 uptake and biogenic respiration estimates not only provide reference data for the estimation of total natural CO2 removal in our study area, but also supply prior input values for the disaggregation of anthropogenic CO2 emissions and biogenic CO2 fluxes when applying top-down approaches to update the megacity’s CO2 emissions inventory. The latter contribution allows unprecedented amounts of GOSAT and ground measurement data regarding CO2 concentration to be analyzed in inverse modeling of anthropogenic CO2 emissions from Tokyo and the Kantō Plain.

Analysis of Spatial and Temporal Variability of the PAR/GHI Ratio and PAR Modeling Based on Two Satellite Estimates

The main objectives of this work are to address the analysis of the spatial and temporal variability of the ratio between photosynthetically active radiation (PAR) and global horizontal irradiance (GHI), as well as to develop PAR models. The analysis was carried out using data from three stations located in mainland Spain covering three climates: oceanic, standard Mediterranean, and continental Mediterranean. The results of this analysis showed a clear dependence between the PAR/GHI ratio and the location; the oceanic climate showed higher values of PAR/GHI compared with Mediterranean climates. Further, the temporal variability of PAR/GHI was conditioned by the variability of clearness index, so it was also higher in oceanic than in Mediterranean climates. On the other hand, Climate Monitoring Satellite Facility (CM-SAF) and Moderate-Resolution Imaging Spectroradiometer (MODIS) data were used to estimate PAR as a function of GHI over the whole territory. The validation with ground measurements showed better performance of the MODIS-estimates-derived model for the oceanic climate (root-mean-square error (RMSE) around 5%), while the model obtained from CM-SAF fitted better for Mediterranean climates (RMSEs around 2%).

Saudi Arabia’s Solar and Wind Energy Penetration: Future Performance and Requirements

Saudi Arabia fully depends on fossil fuels such as oil and natural gas to generate its electricity. Fossil fuels may have limited life and a history of fluctuating costs, which will lead to multiple issues that can affect the energy security of this country in the long-term. Critical Infrastructure Protection (CIP) is a concept different to “energy security”, which must consider the solar and wind energy as basic sources of energy supplies in Saudi Arabia. Monte Carlo Simulation (MCS) and Brownian Motion (BM) approaches were employed to predict the future behaviour of solar and wind energy, along with long-term temperature performance, based on 69 years of historical daily data. MCS and BM were employed to provide a wide range of options for future prediction results. A validation exercise showed that the north-western region was the most highly recommended region for deployment of solar and wind energy applications due to an abundance of solar and wind energy resources with low temperature supported by a clearer sky during the year. This is followed by the southern region, which exhibited good solar and wind energy resources. This study can be considered as a roadmap to meet the climate and sustainability goals by providing a long-term overview of solar energy, wind energy, and temperature performance in some countries that have a lack of long-term future prediction analysis such as Saudi Arabia.

Solar Energy Estimations in India Using Remote Sensing Technologies and Validation with Sun Photometers in Urban Areas

Solar radiation ground data is available in poor spatial resolution, which provides an opportunity and demonstrates the necessity to consider solar irradiance modeling based on satellite data. For the first time, solar energy monitoring in near real-time has been performed for India. This study focused on the assessment of solar irradiance from the Indian Solar Irradiance Operational System (INSIOS) using operational cloud and aerosol data from INSAT-3D and Copernicus Atmosphere Monitoring Service (CAMS)-Monitoring Atmospheric Composition Climate (MACC), respectively. Simulations of the global horizontal irradiance (GHI) and direct normal irradiance (DNI) were evaluated for 1 year for India at four Baseline Surface Radiation Network (BSRN) stations located in urban regions. The INSIOS system outputs as per radiative transfer model results presented high accuracy under clear-sky and cloudy conditions for GHI and DNI. DNI was very sensitive to the presence of cloud and aerosols, where even with small optical depths the DNI became zero, and thus it affected the accuracy of simulations under realistic atmospheric conditions. The median BSRN and INSIOS difference was found to vary from −93 to −49 W/m2 for GHI and −103 to −76 W/m2 for DNI under high solar energy potential conditions. Clouds were able to cause an underestimation of 40%, whereas for various aerosol inputs to the model, the overall accuracy was high for both irradiances, with the coefficient of determination being 0.99, whereas the penetration of photovoltaic installation, which exploits GHI, into urban environments (e.g., rooftop) could be effectively supported by the presented methodology, as estimations were reliable during high solar energy potential conditions. The results showed substantially high errors for monsoon season due to increase in cloud coverage that was not well-predicted at satellite and model resolutions.

Comparison of Artificial Intelligence and Physical Models for Forecasting Photosynthetically-Active Radiation

Different kinds of radiative transfer models, including a relative sunshine-based model (BBM), a physical-based model for tropical environment (PBM), an efficient physical-based model (EPP), a look-up-table-based model (LUT), and six artificial intelligence models (AI) were introduced for modeling the daily photosynthetically-active radiation (PAR, solar radiation at 400–700 nm), using ground observations at twenty-nine stations, in different climatic zones and terrain features, over mainland China. The climate and terrain effects on the PAR estimates from the different PAR models have been quantitatively analyzed. The results showed that the Genetic model had overwhelmingly higher accuracy than the other models, with the lowest root mean square error (RMSE = 0.5 MJ m−2day−1), lowest mean absolute bias error (MAE = 0.326 MJ m−2day−1), and highest correlation coefficient (R = 0.972), respectively. The spatial–temporal variations of the annual mean PAR (APAR), in the different climate zones and terrains over mainland China, were further investigated, using the Genetic model; the PAR values in China were generally higher in summer than those in the other seasons. The Qinghai Tibetan Plateau had always been the area with the highest APAR (8.668 MJ m−2day−1), and the Sichuan Basin had always been the area with lowest APAR (4.733 MJ m−2day−1). The PAR datasets generated by the Genetic model, in this study, could be used in numerous PAR applications, with high accuracy.

Forecasting Root-Zone Electrical Conductivity of Nutrient Solutions in Closed-Loop Soilless Cultures via a Recurrent Neural Network Using Environmental and Cultivation Information

In existing closed-loop soilless cultures, nutrient solutions are controlled by the electrical conductivity (EC) of the solution. However, the EC of nutrient solutions is affected by both growth environments and crop growth, so it is hard to predict the EC of nutrient solution. The objective of this study was to predict the EC of root-zone nutrient solutions in closed-loop soilless cultures using recurrent neural network (RNN). In a test greenhouse with sweet peppers (Capsicum annuum L.), data were measured every 10 s from October 15 to December 31, 2014. Mean values for every hour were analyzed. Validation accuracy (R2) of a single-layer long short-term memory (LSTM) was 0.92 and root-mean-square error (RMSE) was 0.07, which were the best results among the different RNNs. The trained LSTM predicted the substrate EC accurately at all ranges. Test accuracy (R2) was 0.72 and RMSE was 0.08, which were lower than values for the validation. Deep learning algorithms were more accurate when more data were added for training. The addition of other environmental factors or plant growth data would improve model robustness. A trained LSTM can control the nutrient solutions in closed-loop soilless cultures based on predicted future EC. Therefore, the algorithm can make a planned management of nutrient solutions possible, reducing resource waste.

Hourly photosynthetically active radiation estimation in Midwestern United States from artificial neural networks and conventional regressions models

The relationship between hourly photosynthetically active radiation (PAR) and the global solar radiation (R s ) was analyzed from data gathered over 3 years at Bondville, IL, and Sioux Falls, SD, Midwestern USA. These data were used to determine temporal variability of the PAR fraction and its dependence on different sky conditions, which were defined by the clearness index. Meanwhile, models based on artificial neural networks (ANNs) were established for predicting hourly PAR. The performance of the proposed models was compared with four existing conventional regression models in terms of the normalized root mean square error (NRMSE), the coefficient of determination (r (2)), the mean percentage error (MPE), and the relative standard error (RSE). From the overall analysis, it shows that the ANN model can predict PAR accurately, especially for overcast sky and clear sky conditions. Meanwhile, the parameters related to water vapor do not improve the prediction result significantly.

Scientific and technical challenges in remote sensing of plant canopy reflectance and fluorescence

State-of-the-art optical remote sensing of vegetation canopies is reviewed here to stimulate support from laboratory and field plant research. This overview of recent satellite spectral sensors and the methods used to retrieve remotely quantitative biophysical and biochemical characteristics of vegetation canopies shows that there have been substantial advances in optical remote sensing over the past few decades. Nevertheless, adaptation and transfer of currently available fluorometric methods aboard air- and space-borne platforms can help to eliminate errors and uncertainties in recent remote sensing data interpretation. With this perspective, red and blue-green fluorescence emission as measured in the laboratory and field is reviewed. Remotely sensed plant fluorescence signals have the potential to facilitate a better understanding of vegetation photosynthetic dynamics and primary production on a large scale. The review summarizes several scientific challenges that still need to be resolved to achieve operational fluorescence based remote sensing approaches.

Operational application and improvements of the disease risk forecast model PROCULTURE to optimize fungicides spray for the septoria leaf blotch disease in winter wheat in Luxembourg

Abstract. The model PROCULTURE has been developed by the Université Catholique de Louvain – UCL (Belgium) to simulate the progress of the septoria leaf blotch disease on winter wheat during the cropping season. The model has been validated in Luxembourg for four years at four distinct representative sites. It is able to identify infection periods due to the causal agent Mycosphaerella graminicola on the last five leaf layers by combining meteorological data with phenological data from PROCULTURE's crop growth model component. The meteorological forcing consists of hourly time-series of air temperature, relative humidity and cumulative rainfall since the time of sowing, retrieved from automatic weather stations for hindcast and numerical weather prediction model outputs for the forecast periods. In order to improve the model, leaf wetness – which is one of the most important drivers for the spread of the disease – shall be added as an additional predictor. Therefore leaf wetness sensors were set up at four test sites during the 2007 growing season. To get a continuous spatial coverage of the country, it is planned to couple the PROCULTURE model offline to 12-hourly operational weather forecasts from an implementation of the Weather Research and Forecasting (WRF) model for Luxembourg at 1 km resolution. Because the WRF model does not provide leaf wetness directly, an artificial neural network (ANN) is used to model this parameter.

Reconstruction of daily solar UV irradiation from 1893 to 2002 in Potsdam, Germany

Long-term records of solar UV radiation reaching the Earth's surface are scarce. Radiative transfer calculations and statistical models are two options used to reconstruct decadal changes in solar UV radiation from long-term records of measured atmospheric parameters that contain information on the effect of clouds, atmospheric aerosols and ground albedo on UV radiation. Based on earlier studies, where the long-term variation of daily solar UV irradiation was derived from measured global and diffuse irradiation as well as atmospheric ozone by a non-linear regression method [Feister et al. (2002) Photochem Photobiol 76:281-293], we present another approach for the reconstruction of time series of solar UV radiation. An artificial neural network (ANN) was trained with measurements of solar UV irradiation taken at the Meteorological Observatory in Potsdam, Germany, as well as measured parameters with long-term records such as global and diffuse radiation, sunshine duration, horizontal visibility and column ozone. This study is focussed on the reconstruction of daily broad-band UV-B (280-315 nm), UV-A (315-400 nm) and erythemal UV irradiation (ER). Due to the rapid changes in cloudiness at mid-latitude sites, solar UV irradiance exhibits appreciable short-term variability. One of the main advantages of the statistical method is that it uses doses of highly variable input parameters calculated from individual spot measurements taken at short time intervals, which thus do represent the short-term variability of solar irradiance.