High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data
Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate differe...
Ausführliche Beschreibung
Gespeichert in:
| Autor*in: |
Walter T. Dado [verfasserIn] Jillian M. Deines [verfasserIn] Rinkal Patel [verfasserIn] Sang-Zi Liang [verfasserIn] David B. Lobell [verfasserIn] |
|---|
| Format: |
E-Artikel |
|---|---|
| Sprache: |
Englisch |
| Erschienen: |
2020 |
|---|
| Schlagwörter: |
|---|
| Übergeordnetes Werk: |
In: Remote Sensing - MDPI AG, 2009, 12(2020), 21, p 3471 |
|---|---|
| Übergeordnetes Werk: |
volume:12 ; year:2020 ; number:21, p 3471 |
| Links: |
|---|
| DOI / URN: |
10.3390/rs12213471 |
|---|
| Katalog-ID: |
DOAJ047260793 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | DOAJ047260793 | ||
| 003 | DE-627 | ||
| 005 | 20240412211826.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 230227s2020 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.3390/rs12213471 |2 doi | |
| 035 | |a (DE-627)DOAJ047260793 | ||
| 035 | |a (DE-599)DOAJ140b338b733347a9b8c5386512bdd701 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 0 | |a Walter T. Dado |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data |
| 264 | 1 | |c 2020 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 520 | |a Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. | ||
| 650 | 4 | |a crop yields | |
| 650 | 4 | |a yield mapping | |
| 650 | 4 | |a US Corn Belt | |
| 650 | 4 | |a Landsat | |
| 650 | 4 | |a Sentinel | |
| 650 | 4 | |a agricultural monitoring | |
| 653 | 0 | |a Science | |
| 653 | 0 | |a Q | |
| 700 | 0 | |a Jillian M. Deines |e verfasserin |4 aut | |
| 700 | 0 | |a Rinkal Patel |e verfasserin |4 aut | |
| 700 | 0 | |a Sang-Zi Liang |e verfasserin |4 aut | |
| 700 | 0 | |a David B. Lobell |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i In |t Remote Sensing |d MDPI AG, 2009 |g 12(2020), 21, p 3471 |w (DE-627)608937916 |w (DE-600)2513863-7 |x 20724292 |7 nnns |
| 773 | 1 | 8 | |g volume:12 |g year:2020 |g number:21, p 3471 |
| 856 | 4 | 0 | |u https://doi.org/10.3390/rs12213471 |z kostenfrei |
| 856 | 4 | 0 | |u https://doaj.org/article/140b338b733347a9b8c5386512bdd701 |z kostenfrei |
| 856 | 4 | 0 | |u https://www.mdpi.com/2072-4292/12/21/3471 |z kostenfrei |
| 856 | 4 | 2 | |u https://doaj.org/toc/2072-4292 |y Journal toc |z kostenfrei |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a SYSFLAG_A | ||
| 912 | |a GBV_DOAJ | ||
| 912 | |a GBV_ILN_20 | ||
| 912 | |a GBV_ILN_22 | ||
| 912 | |a GBV_ILN_23 | ||
| 912 | |a GBV_ILN_24 | ||
| 912 | |a GBV_ILN_39 | ||
| 912 | |a GBV_ILN_40 | ||
| 912 | |a GBV_ILN_60 | ||
| 912 | |a GBV_ILN_62 | ||
| 912 | |a GBV_ILN_63 | ||
| 912 | |a GBV_ILN_65 | ||
| 912 | |a GBV_ILN_69 | ||
| 912 | |a GBV_ILN_70 | ||
| 912 | |a GBV_ILN_73 | ||
| 912 | |a GBV_ILN_95 | ||
| 912 | |a GBV_ILN_105 | ||
| 912 | |a GBV_ILN_110 | ||
| 912 | |a GBV_ILN_151 | ||
| 912 | |a GBV_ILN_161 | ||
| 912 | |a GBV_ILN_170 | ||
| 912 | |a GBV_ILN_206 | ||
| 912 | |a GBV_ILN_213 | ||
| 912 | |a GBV_ILN_230 | ||
| 912 | |a GBV_ILN_285 | ||
| 912 | |a GBV_ILN_293 | ||
| 912 | |a GBV_ILN_370 | ||
| 912 | |a GBV_ILN_602 | ||
| 912 | |a GBV_ILN_2005 | ||
| 912 | |a GBV_ILN_2009 | ||
| 912 | |a GBV_ILN_2011 | ||
| 912 | |a GBV_ILN_2014 | ||
| 912 | |a GBV_ILN_2055 | ||
| 912 | |a GBV_ILN_2108 | ||
| 912 | |a GBV_ILN_2111 | ||
| 912 | |a GBV_ILN_2119 | ||
| 912 | |a GBV_ILN_4012 | ||
| 912 | |a GBV_ILN_4037 | ||
| 912 | |a GBV_ILN_4112 | ||
| 912 | |a GBV_ILN_4125 | ||
| 912 | |a GBV_ILN_4126 | ||
| 912 | |a GBV_ILN_4249 | ||
| 912 | |a GBV_ILN_4305 | ||
| 912 | |a GBV_ILN_4306 | ||
| 912 | |a GBV_ILN_4307 | ||
| 912 | |a GBV_ILN_4313 | ||
| 912 | |a GBV_ILN_4322 | ||
| 912 | |a GBV_ILN_4323 | ||
| 912 | |a GBV_ILN_4324 | ||
| 912 | |a GBV_ILN_4325 | ||
| 912 | |a GBV_ILN_4335 | ||
| 912 | |a GBV_ILN_4338 | ||
| 912 | |a GBV_ILN_4367 | ||
| 912 | |a GBV_ILN_4392 | ||
| 912 | |a GBV_ILN_4700 | ||
| 951 | |a AR | ||
| 952 | |d 12 |j 2020 |e 21, p 3471 | ||
| author_variant |
w t d wtd j m d jmd r p rp s z l szl d b l dbl |
|---|---|
| matchkey_str |
article:20724292:2020----::iheouinobayedapnarstesiwsuig |
| hierarchy_sort_str |
2020 |
| publishDate |
2020 |
| allfields |
10.3390/rs12213471 doi (DE-627)DOAJ047260793 (DE-599)DOAJ140b338b733347a9b8c5386512bdd701 DE-627 ger DE-627 rakwb eng Walter T. Dado verfasserin aut High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data 2020 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. crop yields yield mapping US Corn Belt Landsat Sentinel agricultural monitoring Science Q Jillian M. Deines verfasserin aut Rinkal Patel verfasserin aut Sang-Zi Liang verfasserin aut David B. Lobell verfasserin aut In Remote Sensing MDPI AG, 2009 12(2020), 21, p 3471 (DE-627)608937916 (DE-600)2513863-7 20724292 nnns volume:12 year:2020 number:21, p 3471 https://doi.org/10.3390/rs12213471 kostenfrei https://doaj.org/article/140b338b733347a9b8c5386512bdd701 kostenfrei https://www.mdpi.com/2072-4292/12/21/3471 kostenfrei https://doaj.org/toc/2072-4292 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_95 GBV_ILN_105 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_370 GBV_ILN_602 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2108 GBV_ILN_2111 GBV_ILN_2119 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4335 GBV_ILN_4338 GBV_ILN_4367 GBV_ILN_4392 GBV_ILN_4700 AR 12 2020 21, p 3471 |
| spelling |
10.3390/rs12213471 doi (DE-627)DOAJ047260793 (DE-599)DOAJ140b338b733347a9b8c5386512bdd701 DE-627 ger DE-627 rakwb eng Walter T. Dado verfasserin aut High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data 2020 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. crop yields yield mapping US Corn Belt Landsat Sentinel agricultural monitoring Science Q Jillian M. Deines verfasserin aut Rinkal Patel verfasserin aut Sang-Zi Liang verfasserin aut David B. Lobell verfasserin aut In Remote Sensing MDPI AG, 2009 12(2020), 21, p 3471 (DE-627)608937916 (DE-600)2513863-7 20724292 nnns volume:12 year:2020 number:21, p 3471 https://doi.org/10.3390/rs12213471 kostenfrei https://doaj.org/article/140b338b733347a9b8c5386512bdd701 kostenfrei https://www.mdpi.com/2072-4292/12/21/3471 kostenfrei https://doaj.org/toc/2072-4292 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_95 GBV_ILN_105 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_370 GBV_ILN_602 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2108 GBV_ILN_2111 GBV_ILN_2119 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4335 GBV_ILN_4338 GBV_ILN_4367 GBV_ILN_4392 GBV_ILN_4700 AR 12 2020 21, p 3471 |
| allfields_unstemmed |
10.3390/rs12213471 doi (DE-627)DOAJ047260793 (DE-599)DOAJ140b338b733347a9b8c5386512bdd701 DE-627 ger DE-627 rakwb eng Walter T. Dado verfasserin aut High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data 2020 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. crop yields yield mapping US Corn Belt Landsat Sentinel agricultural monitoring Science Q Jillian M. Deines verfasserin aut Rinkal Patel verfasserin aut Sang-Zi Liang verfasserin aut David B. Lobell verfasserin aut In Remote Sensing MDPI AG, 2009 12(2020), 21, p 3471 (DE-627)608937916 (DE-600)2513863-7 20724292 nnns volume:12 year:2020 number:21, p 3471 https://doi.org/10.3390/rs12213471 kostenfrei https://doaj.org/article/140b338b733347a9b8c5386512bdd701 kostenfrei https://www.mdpi.com/2072-4292/12/21/3471 kostenfrei https://doaj.org/toc/2072-4292 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_95 GBV_ILN_105 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_370 GBV_ILN_602 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2108 GBV_ILN_2111 GBV_ILN_2119 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4335 GBV_ILN_4338 GBV_ILN_4367 GBV_ILN_4392 GBV_ILN_4700 AR 12 2020 21, p 3471 |
| allfieldsGer |
10.3390/rs12213471 doi (DE-627)DOAJ047260793 (DE-599)DOAJ140b338b733347a9b8c5386512bdd701 DE-627 ger DE-627 rakwb eng Walter T. Dado verfasserin aut High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data 2020 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. crop yields yield mapping US Corn Belt Landsat Sentinel agricultural monitoring Science Q Jillian M. Deines verfasserin aut Rinkal Patel verfasserin aut Sang-Zi Liang verfasserin aut David B. Lobell verfasserin aut In Remote Sensing MDPI AG, 2009 12(2020), 21, p 3471 (DE-627)608937916 (DE-600)2513863-7 20724292 nnns volume:12 year:2020 number:21, p 3471 https://doi.org/10.3390/rs12213471 kostenfrei https://doaj.org/article/140b338b733347a9b8c5386512bdd701 kostenfrei https://www.mdpi.com/2072-4292/12/21/3471 kostenfrei https://doaj.org/toc/2072-4292 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_95 GBV_ILN_105 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_370 GBV_ILN_602 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2108 GBV_ILN_2111 GBV_ILN_2119 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4335 GBV_ILN_4338 GBV_ILN_4367 GBV_ILN_4392 GBV_ILN_4700 AR 12 2020 21, p 3471 |
| allfieldsSound |
10.3390/rs12213471 doi (DE-627)DOAJ047260793 (DE-599)DOAJ140b338b733347a9b8c5386512bdd701 DE-627 ger DE-627 rakwb eng Walter T. Dado verfasserin aut High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data 2020 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. crop yields yield mapping US Corn Belt Landsat Sentinel agricultural monitoring Science Q Jillian M. Deines verfasserin aut Rinkal Patel verfasserin aut Sang-Zi Liang verfasserin aut David B. Lobell verfasserin aut In Remote Sensing MDPI AG, 2009 12(2020), 21, p 3471 (DE-627)608937916 (DE-600)2513863-7 20724292 nnns volume:12 year:2020 number:21, p 3471 https://doi.org/10.3390/rs12213471 kostenfrei https://doaj.org/article/140b338b733347a9b8c5386512bdd701 kostenfrei https://www.mdpi.com/2072-4292/12/21/3471 kostenfrei https://doaj.org/toc/2072-4292 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_95 GBV_ILN_105 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_370 GBV_ILN_602 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2108 GBV_ILN_2111 GBV_ILN_2119 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4335 GBV_ILN_4338 GBV_ILN_4367 GBV_ILN_4392 GBV_ILN_4700 AR 12 2020 21, p 3471 |
| language |
English |
| source |
In Remote Sensing 12(2020), 21, p 3471 volume:12 year:2020 number:21, p 3471 |
| sourceStr |
In Remote Sensing 12(2020), 21, p 3471 volume:12 year:2020 number:21, p 3471 |
| format_phy_str_mv |
Article |
| institution |
findex.gbv.de |
| topic_facet |
crop yields yield mapping US Corn Belt Landsat Sentinel agricultural monitoring Science Q |
| isfreeaccess_bool |
true |
| container_title |
Remote Sensing |
| authorswithroles_txt_mv |
Walter T. Dado @@aut@@ Jillian M. Deines @@aut@@ Rinkal Patel @@aut@@ Sang-Zi Liang @@aut@@ David B. Lobell @@aut@@ |
| publishDateDaySort_date |
2020-01-01T00:00:00Z |
| hierarchy_top_id |
608937916 |
| id |
DOAJ047260793 |
| language_de |
englisch |
| fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">DOAJ047260793</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20240412211826.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">230227s2020 xx |||||o 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.3390/rs12213471</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)DOAJ047260793</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)DOAJ140b338b733347a9b8c5386512bdd701</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="100" ind1="0" ind2=" "><subfield code="a">Walter T. Dado</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2020</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">Computermedien</subfield><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">crop yields</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">yield mapping</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">US Corn Belt</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Landsat</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Sentinel</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">agricultural monitoring</subfield></datafield><datafield tag="653" ind1=" " ind2="0"><subfield code="a">Science</subfield></datafield><datafield tag="653" ind1=" " ind2="0"><subfield code="a">Q</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Jillian M. Deines</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Rinkal Patel</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Sang-Zi Liang</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">David B. Lobell</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">In</subfield><subfield code="t">Remote Sensing</subfield><subfield code="d">MDPI AG, 2009</subfield><subfield code="g">12(2020), 21, p 3471</subfield><subfield code="w">(DE-627)608937916</subfield><subfield code="w">(DE-600)2513863-7</subfield><subfield code="x">20724292</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:12</subfield><subfield code="g">year:2020</subfield><subfield code="g">number:21, p 3471</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.3390/rs12213471</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doaj.org/article/140b338b733347a9b8c5386512bdd701</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://www.mdpi.com/2072-4292/12/21/3471</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="u">https://doaj.org/toc/2072-4292</subfield><subfield code="y">Journal toc</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_DOAJ</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_20</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_22</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_23</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_24</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_39</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_40</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_60</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_62</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_63</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_65</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_69</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_70</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_73</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_95</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_105</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_110</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_151</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_161</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_170</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_206</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_213</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_230</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_285</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_293</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_370</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_602</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2005</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2009</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2011</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2014</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2055</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2108</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2111</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2119</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4012</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4037</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4112</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4125</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4126</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4249</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4305</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4306</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4307</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4313</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4322</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4323</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4324</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4325</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4335</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4338</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4367</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4392</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4700</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">12</subfield><subfield code="j">2020</subfield><subfield code="e">21, p 3471</subfield></datafield></record></collection>
|
| author |
Walter T. Dado |
| spellingShingle |
Walter T. Dado misc crop yields misc yield mapping misc US Corn Belt misc Landsat misc Sentinel misc agricultural monitoring misc Science misc Q High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data |
| authorStr |
Walter T. Dado |
| ppnlink_with_tag_str_mv |
@@773@@(DE-627)608937916 |
| format |
electronic Article |
| delete_txt_mv |
keep |
| author_role |
aut aut aut aut aut |
| collection |
DOAJ |
| remote_str |
true |
| illustrated |
Not Illustrated |
| issn |
20724292 |
| topic_title |
High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data crop yields yield mapping US Corn Belt Landsat Sentinel agricultural monitoring |
| topic |
misc crop yields misc yield mapping misc US Corn Belt misc Landsat misc Sentinel misc agricultural monitoring misc Science misc Q |
| topic_unstemmed |
misc crop yields misc yield mapping misc US Corn Belt misc Landsat misc Sentinel misc agricultural monitoring misc Science misc Q |
| topic_browse |
misc crop yields misc yield mapping misc US Corn Belt misc Landsat misc Sentinel misc agricultural monitoring misc Science misc Q |
| format_facet |
Elektronische Aufsätze Aufsätze Elektronische Ressource |
| format_main_str_mv |
Text Zeitschrift/Artikel |
| carriertype_str_mv |
cr |
| hierarchy_parent_title |
Remote Sensing |
| hierarchy_parent_id |
608937916 |
| hierarchy_top_title |
Remote Sensing |
| isfreeaccess_txt |
true |
| familylinks_str_mv |
(DE-627)608937916 (DE-600)2513863-7 |
| title |
High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data |
| ctrlnum |
(DE-627)DOAJ047260793 (DE-599)DOAJ140b338b733347a9b8c5386512bdd701 |
| title_full |
High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data |
| author_sort |
Walter T. Dado |
| journal |
Remote Sensing |
| journalStr |
Remote Sensing |
| lang_code |
eng |
| isOA_bool |
true |
| recordtype |
marc |
| publishDateSort |
2020 |
| contenttype_str_mv |
txt |
| author_browse |
Walter T. Dado Jillian M. Deines Rinkal Patel Sang-Zi Liang David B. Lobell |
| container_volume |
12 |
| format_se |
Elektronische Aufsätze |
| author-letter |
Walter T. Dado |
| doi_str_mv |
10.3390/rs12213471 |
| author2-role |
verfasserin |
| title_sort |
high-resolution soybean yield mapping across the us midwest using subfield harvester data |
| title_auth |
High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data |
| abstract |
Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. |
| abstractGer |
Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. |
| abstract_unstemmed |
Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model. |
| collection_details |
GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_95 GBV_ILN_105 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_370 GBV_ILN_602 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2108 GBV_ILN_2111 GBV_ILN_2119 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4335 GBV_ILN_4338 GBV_ILN_4367 GBV_ILN_4392 GBV_ILN_4700 |
| container_issue |
21, p 3471 |
| title_short |
High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data |
| url |
https://doi.org/10.3390/rs12213471 https://doaj.org/article/140b338b733347a9b8c5386512bdd701 https://www.mdpi.com/2072-4292/12/21/3471 https://doaj.org/toc/2072-4292 |
| remote_bool |
true |
| author2 |
Jillian M. Deines Rinkal Patel Sang-Zi Liang David B. Lobell |
| author2Str |
Jillian M. Deines Rinkal Patel Sang-Zi Liang David B. Lobell |
| ppnlink |
608937916 |
| mediatype_str_mv |
c |
| isOA_txt |
true |
| hochschulschrift_bool |
false |
| doi_str |
10.3390/rs12213471 |
| up_date |
2024-07-04T00:39:15.288Z |
| _version_ |
1803606876791767040 |
| fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">DOAJ047260793</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20240412211826.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">230227s2020 xx |||||o 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.3390/rs12213471</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)DOAJ047260793</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)DOAJ140b338b733347a9b8c5386512bdd701</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="100" ind1="0" ind2=" "><subfield code="a">Walter T. Dado</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">High-Resolution Soybean Yield Mapping Across the US Midwest Using Subfield Harvester Data</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2020</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">Computermedien</subfield><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Cloud computing and freely available, high-resolution satellite data have enabled recent progress in crop yield mapping at fine scales. However, extensive validation data at a matching resolution remain uncommon or infeasible due to data availability. This has limited the ability to evaluate different yield estimation models and improve understanding of key features useful for yield estimation in both data-rich and data-poor contexts. Here, we assess machine learning models’ capacity for soybean yield prediction using a unique ground-truth dataset of high-resolution (5 m) yield maps generated from combine harvester yield monitor data for over a million field-year observations across the Midwestern United States from 2008 to 2018. First, we compare random forest (RF) implementations, testing a range of feature engineering approaches using Sentinel-2 and Landsat spectral data for 20- and 30-m scale yield prediction. We find that Sentinel-2-based models can explain up to 45% of out-of-sample yield variability from 2017 to 2018 (r<sup<2</sup< = 0.45), while Landsat models explain up to 43% across the longer 2008–2018 period. Using discrete Fourier transforms, or harmonic regressions, to capture soybean phenology improved the Landsat-based model considerably. Second, we compare RF models trained using this ground-truth data to models trained on available county-level statistics. We find that county-level models rely more heavily on just a few predictors, namely August weather covariates (vapor pressure deficit, rainfall, temperature) and July and August near-infrared observations. As a result, county-scale models perform relatively poorly on field-scale validation (r<sup<2</sup< = 0.32), especially for high-yielding fields, but perform similarly to field-scale models when evaluated at the county scale (r<sup<2</sup< = 0.82). Finally, we test whether our findings on variable importance can inform a simple, generalizable framework for regions or time periods beyond ground data availability. To do so, we test improvements to a Scalable Crop Yield Mapper (SCYM) approach that uses crop simulations to train statistical models for yield estimation. Based on findings from our RF models, we employ harmonic regressions to estimate peak vegetation index (VI) and a VI observation 30 days later, with August rainfall as the sole weather covariate in our new SCYM model. Modifications improved SCYM’s explained variance (r<sup<2</sup< = 0.27 at the 30 m scale) and provide a new, parsimonious model.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">crop yields</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">yield mapping</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">US Corn Belt</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Landsat</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Sentinel</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">agricultural monitoring</subfield></datafield><datafield tag="653" ind1=" " ind2="0"><subfield code="a">Science</subfield></datafield><datafield tag="653" ind1=" " ind2="0"><subfield code="a">Q</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Jillian M. Deines</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Rinkal Patel</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Sang-Zi Liang</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">David B. Lobell</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">In</subfield><subfield code="t">Remote Sensing</subfield><subfield code="d">MDPI AG, 2009</subfield><subfield code="g">12(2020), 21, p 3471</subfield><subfield code="w">(DE-627)608937916</subfield><subfield code="w">(DE-600)2513863-7</subfield><subfield code="x">20724292</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:12</subfield><subfield code="g">year:2020</subfield><subfield code="g">number:21, p 3471</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.3390/rs12213471</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doaj.org/article/140b338b733347a9b8c5386512bdd701</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://www.mdpi.com/2072-4292/12/21/3471</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="u">https://doaj.org/toc/2072-4292</subfield><subfield code="y">Journal toc</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_DOAJ</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_20</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_22</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_23</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_24</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_39</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_40</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_60</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_62</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_63</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_65</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_69</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_70</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_73</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_95</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_105</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_110</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_151</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_161</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_170</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_206</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_213</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_230</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_285</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_293</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_370</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_602</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2005</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2009</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2011</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2014</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2055</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2108</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2111</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2119</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4012</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4037</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4112</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4125</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4126</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4249</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4305</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4306</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4307</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4313</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4322</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4323</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4324</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4325</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4335</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4338</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4367</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4392</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4700</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">12</subfield><subfield code="j">2020</subfield><subfield code="e">21, p 3471</subfield></datafield></record></collection>
|
| score |
7.398322 |