TY - JOUR
T1 - Proximal sensing of Urochloa grasses increases selection accuracy
AU - Jiménez, Juan De La Cruz
AU - Leiva, Luisa
AU - Cardoso, Juan A.
AU - French, Andrew N.
AU - Thorp, Kelly R.
PY - 2020/4
Y1 - 2020/4
N2 - In the American tropics, livestock production is highly restricted by forage availability. In addition, the breeding and development of new forage varieties with outstanding yield and high nutritional quality is often limited by a lack of resources and poor technology. Non-destructive, high-throughput phenotyping offers a rapid and economical means of evaluating large numbers of genotypes. In this study, visual assessments, digital colour images, and spectral reflectance data were collected from 200 Urochloa hybrids in a field setting. Partial least-squares regression (PLSR) was applied to relate visual assessments, digital image analysis and spectral data to shoot dry weight, crude protein and chlorophyll concentrations. Visual evaluations of biomass and greenness were collected in 68 min, digital colour imaging data in 40 min, and hyperspectral canopy data in 80 min. Root-mean-squared errors of prediction for PLSR estimations of shoot dry weight, crude protein and chlorophyll were lowest for digital image analysis followed by hyperspectral analysis and visual assessments. This study showed that digital colour image and spectral analysis techniques have the potential to improve precision and reduce time for tropical forage grass phenotyping.
AB - In the American tropics, livestock production is highly restricted by forage availability. In addition, the breeding and development of new forage varieties with outstanding yield and high nutritional quality is often limited by a lack of resources and poor technology. Non-destructive, high-throughput phenotyping offers a rapid and economical means of evaluating large numbers of genotypes. In this study, visual assessments, digital colour images, and spectral reflectance data were collected from 200 Urochloa hybrids in a field setting. Partial least-squares regression (PLSR) was applied to relate visual assessments, digital image analysis and spectral data to shoot dry weight, crude protein and chlorophyll concentrations. Visual evaluations of biomass and greenness were collected in 68 min, digital colour imaging data in 40 min, and hyperspectral canopy data in 80 min. Root-mean-squared errors of prediction for PLSR estimations of shoot dry weight, crude protein and chlorophyll were lowest for digital image analysis followed by hyperspectral analysis and visual assessments. This study showed that digital colour image and spectral analysis techniques have the potential to improve precision and reduce time for tropical forage grass phenotyping.
KW - Brachiaria
KW - phenotyping
KW - plant breeding
KW - tropical forage grasses
UR - http://www.scopus.com/inward/record.url?scp=85083701065&partnerID=8YFLogxK
U2 - 10.1071/CP19324
DO - 10.1071/CP19324
M3 - Article
AN - SCOPUS:85083701065
SN - 1836-0947
VL - 71
SP - 401
EP - 409
JO - Crop and Pasture Science
JF - Crop and Pasture Science
IS - 4
ER -