TY - GEN
T1 - Regression based corn yield assessment using MODIS based daily NDVI in Iowa state
AU - Shrestha, Ranjay
AU - Di, Liping
AU - Yu, Eugene G.
AU - Kang, Lingjun
AU - Li, Lin
AU - Rahman, Md Shahinoor
AU - Deng, Meixia
AU - Yang, Zhengwei
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/9/26
Y1 - 2016/9/26
N2 - Traditional method of visiting field and surveying farmers to estimate crop yield has been considered inefficient and impractical especially in cases when fields are not easily accessible. Remote sensing techniques, therefore, has been utilize to overcome these obstacles with good success. Normalize Difference Vegetation Index (NDVI) based models are considered to be most effective and utilized technique in crop yield assessment and can provide up to field level assessment. This research utilize MODIS based 250m daily NDVI to estimate corn yield in 4 Agricultural Statistics Districts (ASD) in Iowa state. Corn is considered the primary crop in the US accounting for 90% of the total feed grains, hence utilized to be study in this research. Linear regression model was derived between NDVI curve and corn yield using all counties within the 4 ASD between years 2000 to 2014. The regression model showed statistically significant relation between NDVI curve and corn yield with coefficient of deterministic (R-square) over 0.80 in all 4 ASD. Similarly validating the model using new 2015 yield, the average predictability error was between 5 to 7 percent.
AB - Traditional method of visiting field and surveying farmers to estimate crop yield has been considered inefficient and impractical especially in cases when fields are not easily accessible. Remote sensing techniques, therefore, has been utilize to overcome these obstacles with good success. Normalize Difference Vegetation Index (NDVI) based models are considered to be most effective and utilized technique in crop yield assessment and can provide up to field level assessment. This research utilize MODIS based 250m daily NDVI to estimate corn yield in 4 Agricultural Statistics Districts (ASD) in Iowa state. Corn is considered the primary crop in the US accounting for 90% of the total feed grains, hence utilized to be study in this research. Linear regression model was derived between NDVI curve and corn yield using all counties within the 4 ASD between years 2000 to 2014. The regression model showed statistically significant relation between NDVI curve and corn yield with coefficient of deterministic (R-square) over 0.80 in all 4 ASD. Similarly validating the model using new 2015 yield, the average predictability error was between 5 to 7 percent.
KW - Corn Yield
KW - Liner Regression
KW - MODIS NDVI
KW - Remote Sensing
UR - http://www.scopus.com/inward/record.url?scp=84994126094&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84994126094&partnerID=8YFLogxK
U2 - 10.1109/Agro-Geoinformatics.2016.7577657
DO - 10.1109/Agro-Geoinformatics.2016.7577657
M3 - Conference contribution
AN - SCOPUS:84994126094
T3 - 2016 5th International Conference on Agro-Geoinformatics, Agro-Geoinformatics 2016
BT - 2016 5th International Conference on Agro-Geoinformatics, Agro-Geoinformatics 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 5th International Conference on Agro-Geoinformatics, Agro-Geoinformatics 2016
Y2 - 18 July 2016 through 20 July 2016
ER -