TY - JOUR
T1 - Identification of significant factors affecting potato tuber yield for precision management of soil nutrients
AU - Khan, Humna
AU - Esau, Travis J.
AU - Farooque, Aitazaz A.
AU - Abbas, Farhat
AU - Zaman, Qamar Uz
AU - Barrett, Ryan
AU - Acharya, Bishnu
N1 - Funding Information:
This research was supported by the Natural Science and Engineering Research Council of Canada, Prince Edward Island Potato Board, Canadian Horticultural Council, Potatoes New Brunswick, New Brunswick Department of Agriculture, Aquaculture and Fisheries (CAP program), and Agriculture and Agri-Food Canada. The authors are thankful to Joe Brennan and Khalil Al-Mughrabi (New Brunswick Potato Transformation Initiative) and the Precision Agriculture Team at the University of Prince Edward Island for their cooperation and assistance during the experiment.
Publisher Copyright:
© 2021 American Society of Agricultural and Biological Engineers
PY - 2021
Y1 - 2021
N2 - The development of precision agricultural technologies for an adequate potato tuber yield requires knowledge and understanding of significant factors affecting crop growth and its yield. This study was designed to identify the significant factors responsible for fluctuations in tuber yield within the potato fields. The field experiments were conducted during potato growing seasons of 2017 and 2018 in the region of Souris alongside the LaPierre Lane in Prince Edward Island (PEI), Canada. Both fields were under conventional farming and received traditionally recommended agronomic practices and amounts of agrochemicals over the past decade. Grids of 25 x 25 m2 were established to collect horizontal coplanar geometry of ground electrical conductivity (HCP). The soil samples were collected from the same grids to determine soil organic matter content (SOM), pH, Lime Index (LI), phosphorous (P), potassium (K), calcium (Ca), Iron (Fe), cation exchange capacity (CEC) and %P/Al (P aluminum ratio). The soil moisture content (θ) was recorded using time domain reflectometry (TDR) probes. The normalized difference vegetative index (NDVI) was recorded using GreenSeekerTM. Potato tuber yield was collected manually from 0.91 x 3 m2 strips at the same grids. Results revealed that the selected soil and crop properties showed a low, moderate, and high variability and spatial dependence within the fields. Tuber yield had significant (p < 0.001) correlations with HCP, θ, SOM, and P during the first sampling at the beginning of the growing season. However, during the second sampling, the tuber yield was significantly correlated with HCP, θ, NDVI, SOM, P, and K. Stepwise regression (through backward elimination at α = 0.01) excluded the low important variables namely P and K leaving HCP, θ, SOM, and NDVI as the most influential variables for tuber yield. However, these influential variables had significant correlations with soil P and K. Results of this study suggested that managing the crop inputs based on these influential variables has a significant potential to enhance tuber yield.
AB - The development of precision agricultural technologies for an adequate potato tuber yield requires knowledge and understanding of significant factors affecting crop growth and its yield. This study was designed to identify the significant factors responsible for fluctuations in tuber yield within the potato fields. The field experiments were conducted during potato growing seasons of 2017 and 2018 in the region of Souris alongside the LaPierre Lane in Prince Edward Island (PEI), Canada. Both fields were under conventional farming and received traditionally recommended agronomic practices and amounts of agrochemicals over the past decade. Grids of 25 x 25 m2 were established to collect horizontal coplanar geometry of ground electrical conductivity (HCP). The soil samples were collected from the same grids to determine soil organic matter content (SOM), pH, Lime Index (LI), phosphorous (P), potassium (K), calcium (Ca), Iron (Fe), cation exchange capacity (CEC) and %P/Al (P aluminum ratio). The soil moisture content (θ) was recorded using time domain reflectometry (TDR) probes. The normalized difference vegetative index (NDVI) was recorded using GreenSeekerTM. Potato tuber yield was collected manually from 0.91 x 3 m2 strips at the same grids. Results revealed that the selected soil and crop properties showed a low, moderate, and high variability and spatial dependence within the fields. Tuber yield had significant (p < 0.001) correlations with HCP, θ, SOM, and P during the first sampling at the beginning of the growing season. However, during the second sampling, the tuber yield was significantly correlated with HCP, θ, NDVI, SOM, P, and K. Stepwise regression (through backward elimination at α = 0.01) excluded the low important variables namely P and K leaving HCP, θ, SOM, and NDVI as the most influential variables for tuber yield. However, these influential variables had significant correlations with soil P and K. Results of this study suggested that managing the crop inputs based on these influential variables has a significant potential to enhance tuber yield.
KW - Precision agriculture
KW - Regression analysis
KW - Sensors
KW - Site-specific nutrient management
KW - Spatial variability
UR - http://www.scopus.com/inward/record.url?scp=85107930908&partnerID=8YFLogxK
U2 - 10.13031/aea.14127
DO - 10.13031/aea.14127
M3 - Article
AN - SCOPUS:85107930908
SN - 0883-8542
VL - 37
SP - 535
EP - 545
JO - Applied Engineering in Agriculture
JF - Applied Engineering in Agriculture
IS - 3
ER -