Developing prediction models for biomass components of Tectona grandis L.
Loading...
Date
2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Punjab Agricultural University, Ludhiana
Abstract
The present investigation entitled ‘DEVELOPING PREDICTION MODELS FOR BIOMASS
COMPONENTS OF Tectona grandis L’ was carried out at the University Seed Farm, Ladhowal,
Punjab Agricultural University (PAU), Ludhiana during the year 2015-16. Tectona grandis
belonging to family Verbenaceae is widely distributed and economically important species for
timber and furniture. Species is being planted at many places in Punjab and many farmers are
showing interest in the species. Keeping in view of its encouraging response under Punjab
conditions, so for no biomass prediction models/equations are available to ascertain the growing
stock, which will be a usually handy tool to know the biomass. The proposed study was a step
formed in this direction. Among many models developed by using different tree growth
parameters [Height (H), girth at breast height (GBH) and GBH2×H)], the best fit equations were
found with Linear and exponential regression functions using GBH and GBH2×H parameters as
independent factor with various dependent factors [total green and dry biomass (TGB & TDB),
total above and below ground green (AGB & BGB) and dry biomass (ADB & BDB)]. Among
developed models, the logarithm function i.e. TGB = 240.794ln GBH - 668.336 and quadratic
equation TGB = 0.029GBH2+7.853GBH-53.814 with approximately the same Adj. R2 0.94 and
0.92, respectively were found best fit lines to predict total green and dry biomass of tree.
Similarly, for above ground green and dry biomass, logarithm function AGB=204.418lnGBH591.945
with Adj. R2 value of 0.94 and quadratic function AGB = -0.001GBH2H+5.195GBH26.301
with Adj. R2 value of 0.93 both were found best fit to predict above ground biomass and
for below dry ground green and biomass prediction. The power function BFB = 3.257GBH(0.714)
with maximum Adj. R2 value of 0.71 followed by exponential function BFB = 29.668e(0.011GBH)
with Adj. R2 value of 0.67 and logarithm function BFB = 43.243lnGBH-112.919 with minimum
Adj. R2 value of 0.66 were significantly found best fit. Even the regression coefficient of the
logarithm and quadratic function were very high but in validation these models shows less
reliability in prediction for higher and lower GBH class components so we give preference to
linear and exponential functions due to there more compatibility with all components in all GBH
classes. Biomass expansion factor (BEF) showed decreasing trend as the range of the girth classes
increased. For different GBH classes, the values of BEF ranged from 1.61 to 1.24 and with
increase in girth class.
Description
Keywords
null