Abstract:
In the UK, coppice willow is a potentially high yielding energy crop which if widely
planted could have significant effects of hydrology. This is one reason why researchers
are interested in developing reliable techniques for estimating whole-plant water use of
such trees. In recent years various sap flow measuring techniques have become
commercially available. The aim of this thesis was to evaluate the reliability sap flow
measurement and to consider issues related to best practice when used on multi-stemmed woody plants. The thesis starts by reviewing methodologies for estimating
tree water-use with particular focus on sap flow gauges. Subsequently over three years,
experiments were undertaken using a commercially available Stem Heat Balance (SHB)
sap flow gauges, manufactured by Dynamax Inc, Houston, Texas, on coppice willow
grown in lysimeters. Plant responses were monitored through a range of soil-water
conditions from flood to drought. Over three years a methodology for deriving an
accurate assessment of total plant water-use using a lysimeter water balance (LWB) was
developed. Whole-plant water-use, estimated from scaled up sap flow measurements
from individual stems were compared against LWB values. Both, stem basal area and
leaf area were used as scalars to derive values of plant water-use.
In the final experiment, four out of eight, different sized sap flow gauges, with the
appropriate scalar, gave estimates similar (±7%) to LWB values of whole-plant water use over a period of ‘unstressed’ growing conditions. Variation in the accuracy of
estimates was considered to be a function of a) error inherent to the SHB technique, b)
error in scalar values used to derive whole-plant estimates, and c) apparently
autonomous responses of individual stems to changes in soil water status. In non-water
stressed conditions and where sap flow rates are high, errors from the technique were
minimised by selecting Targe’ stems (16-19 mm diameter). Under extreme water stress
conditions, reductions in leaf area can result in errors if the estimate is based on stem diameter. Where individual stem flow rates were perturbed by changes in soil-water
conditions, selecting an ‘intermediate’ sized stem (in this case 10-13 mm diameter)
appeared to minimise errors.
