ESTIMATING LEAF MASS FROM LIGHT INTERCEPTION MEASUREMENTS ON ISOLATEDPLANTS OF ERYTHROXYLUM SPECIES

Citation
Mc. Acock et al., ESTIMATING LEAF MASS FROM LIGHT INTERCEPTION MEASUREMENTS ON ISOLATEDPLANTS OF ERYTHROXYLUM SPECIES, Agronomy journal, 86(3), 1994, pp. 570-574
Citations number
11
Categorie Soggetti
Agriculture
Journal title
ISSN journal
00021962
Volume
86
Issue
3
Year of publication
1994
Pages
570 - 574
Database
ISI
SICI code
0002-1962(1994)86:3<570:ELMFLI>2.0.ZU;2-X
Abstract
To determine the extent of coca (Eryfhroxylum spp.) production in the world, some method for estimating yield (leaf mass) in a variety of ha bitats is required. A computer model that simulates the growth of coca plants has the versatitity for making these estimates, but there must be some reliable method to validate the model. The objective of this research was to validate the use of measurements of light interception and specific leaf area (SLA, m(2) kg(-1)) for estimating yield. Leaf area was calculated from light interception measurements made on plant s grown as hedgerow or as isolated plants, then converted to leaf mass using measurements of SLA. Light interception was measured with the D ecagon Sunfleck Ceptometer (SC) and the LI-COR Plant Canopy Analyzer ( PCA). Leaf disks of known area from Erythroxylum coca var. coca and Er ythroxylum novogranatense var. novogranatense were cut, dried, and wei ghed to determine SLA values. Leaf fresh weights were measured and cor rected to dry weight using the fresh to dry weight ratio obtained from a sample at harvest. Estimated leaf mass regressed on observed leaf m ass demonstrated that PCA was superior to SC in estimating leaf mass o n isolated plants of various sizes using the techniques described in t his paper. Leaf area was consistently underestimated by the SC. The es timates of leaf mass from the PCA gave the lower mean, absolute mean, and root mean square error terms compared with SC, and the slope of th e regression line of observed leaf mass vs. predicted leaf mass was cl osest to a value of 1.0.