Spatio-temporal downscaling of gridded crop model yield estimates based on machine learning

15505 13 85323 3 239672 document 85323 Folberth_et_al_AGFORMET_2019.pdf application/pdf 0fcd1c934cdd2f5a7bc0be35840b263d MD5 2126867 2018-10-05 11:15:32 https://pure.iiasa.ac.at/id/eprint/15505/1/Folberth_et_al_AGFORMET_2019.pdf 15505 1 1 application/pdf text en public cc_by_nc_nd_4

Folberth_et_al_AGFORMET_2019.pdf

accepted 85324 2 239675 document 85324 lightbox.jpg image/png cdfcfeeca9307bda43701b63e112828a MD5 23200 2018-10-05 11:25:50 https://pure.iiasa.ac.at/id/eprint/15505/2/lightbox.jpg 15505 2 2 image/png other Thumbnails conversion from text to thumbnail_lightbox en staffonly

lightbox.jpg

preview.jpg

medium.jpg

small.jpg

indexcodes.txt

http://eprints.org/relation/isVersionOf https://pure.iiasa.ac.at/id/document/85323 http://eprints.org/relation/isVolatileVersionOf https://pure.iiasa.ac.at/id/document/85323 http://eprints.org/relation/isIndexCodesVersionOf https://pure.iiasa.ac.at/id/document/85323 archive 353 disk0/00/01/55/05 2018-10-05 10:07:40 2021-08-27 17:30:51 2018-10-05 10:07:40 article show Folberth C. 2033 0000-0002-6738-5238 Baklanov A. 2067 0000-0003-1599-3618 Balkovic J. 1977 0000-0003-2955-4931 Skalsky R. 1986 0000-0002-0983-6897 Khabarov N. 1850 0000-0001-5372-4668 Obersteiner M. 7255 0000-0001-6981-2769 Spatio-temporal downscaling of gridded crop model yield estimates based on machine learning pub prog_asa prog_esm Meta-model; Extreme gradient boosting; Random forests; Maize yield; Agricultural externalities; Climate features Global gridded crop models (GGCMs) are essential tools for estimating agricultural crop yields and externalities at large scales, typically at coarse spatial resolutions. Higher resolution estimates are required for robust agricultural assessments at regional and local scales, where the applicability of GGCMs is often limited by low data availability and high computational demand. An approach to bridge this gap is the application of meta-models trained on GGCM output data to covariates of high spatial resolution. In this study, we explore two machine learning approaches – extreme gradient boosting and random forests - to develop meta-models for the prediction of crop model outputs at fine spatial resolutions. Machine learning algorithms are trained on global scale maize simulations of a GGCM and exemplary applied to the extent of Mexico at a finer spatial resolution. Results show very high accuracy with R2>0.96 for predictions of maize yields as well as the hydrologic externalities evapotranspiration and crop available water with also low mean bias in all cases. While limited sets of covariates such as annual climate data alone provide satisfactory results already, a comprehensive set of predictors covering annual, growing season, and monthly climate data is required to obtain high performance in reproducing climate-driven inter-annual crop yield variability. The findings presented herein provide a first proof of concept that machine learning methods are highly suitable for building crop meta-models for spatio-temporal downscaling and indicate potential for further developments towards scalable crop model emulators. 2019-01-15 published Elsevier 10.1016/j.agrformet.2018.09.021 92 22 23 287 151 219 public Agricultural and Forest Meteorology 264 1-15 TRUE 1873-2240 http://pure.iiasa.ac.at/id/eprint/15490/ data Effects of phosphorus limitations on Life, Earth system and Society (IMBALANCE-P, FP7 610028) FALSE yes info:eu-repo/grantAgreement/EC/FP7/610028/EU/Effects of phosphorus limitations on Life, Earth system and Society/IMBALANCE-P info:eu-repo/semantics/article info:eu-repo/semantics/closedAccess