scholarly journals Towards Predictive Modeling of Sorghum Biomass Yields Using Fraction of Absorbed Photosynthetically Active Radiation Derived from Sentinel-2 Satellite Imagery and Supervised Machine Learning Techniques

Agronomy ◽  
2019 ◽  
Vol 9 (4) ◽  
pp. 203 ◽  
Author(s):  
Ephrem Habyarimana ◽  
Isabelle Piccard ◽  
Marcello Catellani ◽  
Paolo De Franceschi ◽  
Michela Dall’Agata
Keyword(s):  
Machine Learning ◽  
Predictive Modeling ◽  
Aboveground Biomass ◽  
Photosynthetically Active Radiation ◽  
Supervised Machine Learning ◽  
Policy Makers ◽  
Active Radiation ◽  
Biomass Yields ◽  
Absorbed Photosynthetically Active Radiation ◽  
Sentinel 2

Sorghum crop is grown under tropical and temperate latitudes for several purposes including production of health promoting food from the kernel and forage and biofuels from aboveground biomass. One of the concerns of policy-makers and sorghum growers is to cost-effectively predict biomass yields early during the cropping season to improve biomass and biofuel management. The objective of this study was to investigate if Sentinel-2 satellite images could be used to predict within-season biomass sorghum yields in the Mediterranean region. Thirteen machine learning algorithms were tested on fortnightly Sentinel-2A and Sentinel-2B estimates of the fraction of Absorbed Photosynthetically Active Radiation (fAPAR) in combination with in situ aboveground biomass yields from demonstrative fields in Italy. A gradient boosting algorithm implementing the xgbtree method was the best predictive model as it was satisfactorily implemented anywhere from May to July. The best prediction time was the month of May followed by May–June and May–July. To the best of our knowledge, this work represents the first time Sentinel-2-derived fAPAR is used in sorghum biomass predictive modeling. The results from this study will help farmers improve their sorghum biomass business operations and policy-makers and extension services improve energy planning and avoid energy-related crises.

scholarly journals Upscaling from Instantaneous to Daily Fraction of Absorbed Photosynthetically Active Radiation (FAPAR) for Satellite Products

2020 ◽  
Vol 12 (13) ◽  
pp. 2083
Author(s):  
Siyuan Chen ◽  
Liangyun Liu ◽  
Xue He ◽  
Zhigang Liu ◽  
Dailiang Peng
Keyword(s):  
Zenith Angle ◽  
Photosynthetically Active Radiation ◽  
Solar Zenith Angle ◽  
Field Measurements ◽  
Area Index ◽  
Active Radiation ◽  
Sensing Applications ◽  
Typical Satellite ◽  
Absorbed Photosynthetically Active Radiation ◽  
Local Noon

The fraction of absorbed photosynthetically active radiation (FAPAR) is an essential climate variable (ECV) widely used for various ecological and climate models. However, all the current FAPAR satellite products correspond to instantaneous FAPAR values acquired at the satellite transit time only, which cannot represent the variations in photosynthetic processes over the diurnal period. Most studies have directly used the instantaneous FAPAR as a reasonable approximation of the daily integrated value. However, clearly, FAPAR varies a lot according to the weather conditions and amount of incoming radiation. In this paper, a temporal upscaling method based on the cosine of the solar zenith angle (SZA) at local noon ( c o s ( S Z A n o o n ) ) is proposed for converting instantaneous FAPAR to daily integrated FAPAR. First, the diurnal variations in FAPAR were investigated using PROSAIL (a model of Leaf Optical Properties Spectra (PROSPECT) integrating a canopy radiative transfer model (Scattering from Arbitrarily Inclined Leaves, SAIL)) simulations with different leaf area index (LAI) values corresponding to different latitudes. It was found that the instantaneous black sky FAPAR at 09:30 AM provided a good approximation for the daily integrated black sky FAPAR; this gave the highest correlation (R2 = 0.995) and lowest Root Mean Square Error (RMSE = 0.013) among the instantaneous black sky FAPAR values observed at different times. Secondly, the difference between the instantaneous black sky FAPAR values acquired at different times and the daily integrated black sky FAPAR was analyzed; this could be accurately modelled using the cosine value of solar zenith angle at local noon ( c o s ( S Z A n o o n ) ) for a given vegetation scene. Therefore, a temporal upscaling method for typical satellite products was proposed using a cos(SZA)-based upscaling model. Finally, the proposed cos(SZA)-based upscaling model was validated using both the PROSAIL simulated data and the field measurements. The validated results indicated that the upscaled daily black sky FAPAR was highly consistent with the daily integrated black sky FAPAR, giving very high mean R2 values (0.998, 0.972), low RMSEs (0.007, 0.014), and low rMAEs (0.596%, 1.378%) for the simulations and the field measurements, respectively. Consequently, the cos(SZA)-based method performs well for upscaling the instantaneous black sky FAPAR to its daily value, which is a simple but extremely important approach for satellite remote sensing applications related to FAPAR.

Absorbed photosynthetically active radiation and sun‐view geometry effects on remote sensing relationships∗

1998 ◽  
Vol 17 (1-4) ◽  
pp. 89-102 ◽  
Author(s):  
E.A. Walter‐Shea ◽  
B.L. Blad ◽  
M.A. Mesarch ◽  
C.J. Hays ◽  
D.W. Deering ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Photosynthetically Active Radiation ◽  
Active Radiation ◽  
Geometry Effects ◽  
Absorbed Photosynthetically Active Radiation
Sign in / Sign up

Export Citation Format

Share Document