Subfield-level crop yield mapping without ground truth data: A scale transfer framework

IF 11.1 1区 地球科学 Q1 ENVIRONMENTAL SCIENCES Remote Sensing of Environment Pub Date : 2024-09-13 DOI:10.1016/j.rse.2024.114427
{"title":"Subfield-level crop yield mapping without ground truth data: A scale transfer framework","authors":"","doi":"10.1016/j.rse.2024.114427","DOIUrl":null,"url":null,"abstract":"<div><p>Ongoing advances in satellite remote sensing data and machine learning methods have enabled crop yield estimation at various spatial and temporal resolutions. While yield mapping at broader scales (e.g., state or county level) has become common, mapping at finer scales (e.g., field or subfield) has been limited by the lack of ground truth data for model training and evaluation. Here we present a scale transfer framework, named Quantile loss Domain Adversarial Neural Networks (QDANN), that leverages knowledge from county-level datasets to map crop yields at the subfield level. Based on the strategy of unsupervised domain adaptation, QDANN is trained on labeled county-level data and unlabeled subfield-level data, with no requirement for yield information at the subfield level. We evaluate the proposed method applied to Landsat imagery and Gridmet weather data for maize, soybean, and winter wheat fields in the United States, using as reference data yield monitor records from roughly one million field-year observations. The model is compared with several process-based and machine learning-based benchmark approaches that train on simulated yield records or county-level data. QDANN-estimated yields achieved an R<sup>2</sup> score (RMSE) of 48 % (2.29 t/ha), 32 % (0.85 t/ha), and 39 % (1.40 t/ha) for maize, soybean, and winter wheat in comparison with the ground-based yield measures, respectively. These performances are higher than benchmark approaches and are nearly as good as models trained on field-level data. When aggregated to the county level, the improvement achieved by QDANN is more pronounced and the R<sup>2</sup> scores (RMSE) improved to 78 % (0.98 t/ha), 62 % (0.37 t/ha), and 53 % (1.00 t/ha) for maize, soybean, and winter wheat, respectively. This study demonstrates that the proposed scale transfer framework can serve as a reliable approach for yield mapping at the subfield level when there is no access to fine-scale yield information. Based on the QDANN model, we have generated and made publicly available 30-m annual yield maps for major crop-producing states in the U.S. since 2008.</p></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":null,"pages":null},"PeriodicalIF":11.1000,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Remote Sensing of Environment","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S003442572400453X","RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
引用次数: 0

Abstract

Ongoing advances in satellite remote sensing data and machine learning methods have enabled crop yield estimation at various spatial and temporal resolutions. While yield mapping at broader scales (e.g., state or county level) has become common, mapping at finer scales (e.g., field or subfield) has been limited by the lack of ground truth data for model training and evaluation. Here we present a scale transfer framework, named Quantile loss Domain Adversarial Neural Networks (QDANN), that leverages knowledge from county-level datasets to map crop yields at the subfield level. Based on the strategy of unsupervised domain adaptation, QDANN is trained on labeled county-level data and unlabeled subfield-level data, with no requirement for yield information at the subfield level. We evaluate the proposed method applied to Landsat imagery and Gridmet weather data for maize, soybean, and winter wheat fields in the United States, using as reference data yield monitor records from roughly one million field-year observations. The model is compared with several process-based and machine learning-based benchmark approaches that train on simulated yield records or county-level data. QDANN-estimated yields achieved an R2 score (RMSE) of 48 % (2.29 t/ha), 32 % (0.85 t/ha), and 39 % (1.40 t/ha) for maize, soybean, and winter wheat in comparison with the ground-based yield measures, respectively. These performances are higher than benchmark approaches and are nearly as good as models trained on field-level data. When aggregated to the county level, the improvement achieved by QDANN is more pronounced and the R2 scores (RMSE) improved to 78 % (0.98 t/ha), 62 % (0.37 t/ha), and 53 % (1.00 t/ha) for maize, soybean, and winter wheat, respectively. This study demonstrates that the proposed scale transfer framework can serve as a reliable approach for yield mapping at the subfield level when there is no access to fine-scale yield information. Based on the QDANN model, we have generated and made publicly available 30-m annual yield maps for major crop-producing states in the U.S. since 2008.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
在没有地面实况数据的情况下绘制子田级作物产量图:规模转移框架
卫星遥感数据和机器学习方法的不断进步使各种空间和时间分辨率的作物产量估算成为可能。虽然更广尺度(如州或县级)的产量测绘已很普遍,但由于缺乏用于模型训练和评估的地面实况数据,更细尺度(如田间或分田)的测绘一直受到限制。在此,我们提出了一个规模转移框架,名为 "量子损失域对抗神经网络(QDANN)",该框架可利用县级数据集的知识绘制子田级别的作物产量图。基于无监督域适应策略,QDANN 在有标记的县级数据和无标记的子田级数据上进行训练,对子田级的产量信息没有要求。我们评估了应用于 Landsat 图像和美国玉米、大豆和冬小麦田 Gridmet 气象数据的建议方法,并将大约一百万个田间年观测的产量监测记录作为参考数据。该模型与基于模拟产量记录或县级数据进行训练的几种基于过程和机器学习的基准方法进行了比较。与地面测产结果相比,QDANN 估算的玉米、大豆和冬小麦产量的 R2 值(RMSE)分别为 48%(2.29 吨/公顷)、32%(0.85 吨/公顷)和 39%(1.40 吨/公顷)。这些性能均高于基准方法,几乎与基于田间数据训练的模型相当。当汇总到县一级时,QDANN 的改进更为明显,玉米、大豆和冬小麦的 R2 分数(均方根误差)分别提高到 78 %(0.98 吨/公顷)、62 %(0.37 吨/公顷)和 53 %(1.00 吨/公顷)。这项研究表明,在无法获得精细尺度产量信息的情况下,所提出的尺度转移框架可作为一种可靠的方法,用于子田水平的产量测绘。基于 QDANN 模型,我们自 2008 年起生成并公开了美国主要作物生产州的 30 米年产量图。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Remote Sensing of Environment
Remote Sensing of Environment 环境科学-成像科学与照相技术
CiteScore
25.10
自引率
8.90%
发文量
455
审稿时长
53 days
期刊介绍: Remote Sensing of Environment (RSE) serves the Earth observation community by disseminating results on the theory, science, applications, and technology that contribute to advancing the field of remote sensing. With a thoroughly interdisciplinary approach, RSE encompasses terrestrial, oceanic, and atmospheric sensing. The journal emphasizes biophysical and quantitative approaches to remote sensing at local to global scales, covering a diverse range of applications and techniques. RSE serves as a vital platform for the exchange of knowledge and advancements in the dynamic field of remote sensing.
期刊最新文献
Sensitivity of Sentinel-1 C-band SAR backscatter, polarimetry and interferometry to snow accumulation in the Alps Impact of altimeter-buoy data-pairing methods on the validation of Sentinel-3A coastal significant wave heights An in situ approach for validation of canopy chlorophyll fluorescence radiative transfer models using the full emission spectrum Assessment of an Adaptive Subwaveform Coastal Retracker (ASCR) over global coastal oceans for SAR altimetry Exploring the potential of SAR and terrestrial and airborne LiDAR in predicting forest floor spectral properties in temperate and boreal forests
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1