Predicting spatiotemporal patterns of productivity and grazing from multispectral data using neural network analysis based on system complexity

Author:

Ashworth A. J.1ORCID,Avila A.2,Smith H.3ORCID,Winzeler T. E.2,Owens P.4ORCID,Flynn C.5,O'Brien P.6ORCID,Philipp D.7,Su J.2

Affiliation:

1. USDA‐ARS, Poultry Production and Product Safety Research Unit Fayetteville Arkansas USA

2. Department of Mathematics University of Texas at Arlington Arlington Texas USA

3. Environmental Dynamics Program University of Arkansas Fayetteville Arkansas USA

4. USDA‐ARS, Dale Bumpers Small Farms Research Center Booneville Arkansas USA

5. Grassland Soil and Water Research Laboratory Temple Texas USA

6. USDA ARS National Laboratory for Agriculture and The Environment Ames Iowa USA

7. Animal Science Department University of Arkansas Fayetteville Arkansas USA

Abstract

AbstractRemote sensing tools, along with Global Navigation Satellite System cattle collars and digital soil maps, may help elucidate spatiotemporal relationships among soils, terrain, forages, and animals. However, standard computational procedures preclude systems‐level evaluations across this continuum due to data inoperability and processing limitations. Deep learning, a subset of neural network, may elucidate efficiency of livestock production and linkages within the livestock‐grazing environment. Consequently, we applied deep learning to environmental remote sensing data to (1) develop predictive models for yield and forage nutrition based on vegetation indices and (2) at a pixel‐level (per 55 m2), identify how grazing is linked to soil properties, forage growth and nutrition, and terrain attributes in silvopasture and pasture‐only systems. Remotely sensed data rapidly and non‐destructively estimated herbage mass and nutritive value for enhanced net and primary productivity management in livestock and grazing systems. Cattle grazed big bluestem (Andropogon gerardii ‘Vitman’) with 182% greater frequency than orchardgrass (Dactylis glomerata L.) in the pasture‐only system. Real‐time estimates of vegetative bands may assist in predicting grazing pressure for more efficient pasture resource management. Cattle grazing followed distinct soil‐landscape patterns, namely reduced cattle grazing preference occurred in areas of water accumulation, which highlights linkages among terrain, soil‐water movement, soil properties, forage nutrition, and animal grazing response spatially and temporally. Results from this study could be scaled up to improve grazing management among the largest land‐use category in the United States, that is, grasslands, which would allow for sustainable intensification of forage‐based livestock production to meet growing demands for environmentally responsible protein.

Publisher

Wiley

Reference66 articles.

1. Abadi M. Agarwal A. Barham P. Brevdo E. Chen Z. Citro C. Corrado G. S. Davis A. Dean J. Devin M. Ghemawat S. Goodfellow I. Harp A. Irving G. Isard M. Jozefowicz R. Jia Y. Kaiser L. Kudlur M. …Zheng X.(2015).TensorFlow: Large‐scale machine learning on heterogeneous systems(Version v2) [Computer software]. arXiv.https://doi.org/10.48550/arXiv.1603.04467

2. Soil CO 2 evolution is driven by forage species, soil moisture, grazing pressure, poultry litter fertilization, and seasonality in silvopastures

3. Relating topography and soil phosphorus distribution in litter‐amended pastures in Arkansas

4. Topographic Controls on Soil Nutrient Variations in a Silvopasture System

5. Significance of epochs on training a neural network;Afaq S.;International Journal of Scientific & Technology Research,2020

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3