USING SPECTROSCOPIC DATA SETS TO PREDICT NUMBERS OF THRIPS (FULMEKIOLA SERRATA) IN SUGARCANE
USING SPECTROSCOPIC DATA SETS TO PREDICT NUMBERS OF THRIPS (FULMEKIOLA SERRATA) IN SUGARCANE
No Thumbnail Available
Date
2009
Authors
Abdel-Rahman, Elfatih
Journal Title
Journal ISSN
Volume Title
Publisher
uofk
Abstract
Sugarcane thrips, Fulmekiola serrata (Kobus) (Thysanoptera: Thripidae), has recently been
perceived as a constraint to southern African sugarcane production. Spectroscopic readings at
leaf level in sugarcane variety N19 were recorded with the aim of predicting numbers of
thrips (nymphs, adults and nymphs+adults) using partial least squares (PLS) regression.
Correlations were high, viz. R2=0.75 for predicting nymphs numbers, 0.72 for adults and 0.75
for nymphs+adults. Predictions were reliable for nymphs since root mean square error
(RMSE) of prediction was 1.25 (47%). Higher RMSE values were obtained for adults (4.8;
66%) and combined life stages (6.2; 63%).
Description
This paper had been presented for promotion at the university of Khartoum. To get the full text please contact the author
elfatihabdelrahman@gmail.com
Keywords
sugarcane, thrips, Fulmekiola serrata, spectroscopy, partial least squares regression