Premium
Estimation of methane emissions based on crop yield and remote sensing data in a paddy field
Author(s) -
Shi Yifan,
Lou Yunsheng,
Zhang Zhen,
Ma Li,
Ojara Moses A
Publication year - 2020
Publication title -
greenhouse gases: science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.45
H-Index - 32
ISSN - 2152-3878
DOI - 10.1002/ghg.1946
Subject(s) - environmental science , biomass (ecology) , normalized difference vegetation index , hyperspectral imaging , enhanced vegetation index , mean squared error , vegetation (pathology) , greenhouse gas , yield (engineering) , paddy field , canopy , growing season , vegetation index , leaf area index , mathematics , agronomy , remote sensing , statistics , ecology , geography , materials science , metallurgy , biology , medicine , pathology
Quantifying agricultural greenhouse gas (GHG) emissions is important for addressing global warming. In this regard, empirical models were constructed to evaluate the feasibility of using rice yield and canopy spectral properties for estimating paddy cumulative methane (CH 4 ) emissions (CCE). A field experiment with shading treatments was conducted in 2017. A static chamber‐gas chromatography was used to measure CH 4 fluxes during the growing season. Canopy hyperspectral reflectance was measured and then used to calculate the multispectral normalized difference vegetation index (NDVI), the ratio vegetation index (RVI), and the enhanced vegetation index (EVI). The results show that CH 4 emissions were positively correlated to rice yield and biomass, indicating that higher biomass provided more substrates for CH 4 generation. O 2 availability maybe a main factor in the CCE differences under shading treatments. RVI and EVI showed a stronger positive relationship to CH 4 emissions than NDVI. The empirical model including yield and EVI‐JS (EVI of jointing‐booting stage) as input variables performed better (adj‐ R 2 = 0.85, root mean square error [RMSE] = 12.26 kg ha −1 ) than the model that included only yield (adj‐ R 2 = 0.5, RMSE = 22.5 kg ha −1 ). This study suggests that it is feasible to apply remote sensing in paddy CH 4 estimation, providing a referable attempt for future regional agricultural GHG emissions quantification. © 2020 Society of Chemical Industry and John Wiley & Sons, Ltd.