Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.464101
Title: A comparison of physiographic analysis with conventional soil surveys for forest site classification in north east Scotland
Author: McGarry, D.
Awarding Body: University of Aberdeen
Current Institution: University of Aberdeen
Date of Award: 1979
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Please try the link below.
Access through Institution:
Abstract:
The objective of the project is to determine which survey technique provides the most practicable and efficient basis for forest site classification. Prior knowledge of the potential growth rates of Sitka spruce and of the management inputs required for various soil and physiographic units is of paramount importance if an accurate survey of previously unplanted hill land is to be made. Such information would provide the basis for a scientifically based and meaningful land capability for forestry. The soils of two Forestry Commission forests (Kirkhill and Bennachie) with a total area of 560 ha were" surveyed following the Forestry Commission Site Survey methods and classification system. For the same two areas physiographic units (facets) were drawn on air photographs. The two areas have also been surveyed by the National Soil Survey. In this way three systems of land survey were available for the two forested areas. To test the homogeneity of the three sets of survey units 158 plots (0.01 ha circles) were located over both forests and within the three sets of classification units, each plot being a random point-sample of each of the three survey systems. Plot data collected were: various depth measures, texture, percent and size of stones in the profile; altitude, slope and aspect; age, topheight, diameter and five year intercept of trees; soil samples for chemical analyses. Initially, coefficients of variation and pair-wise t-tests of survey unit means were used to examine the plot data within each forest, and two-way analysis of variance to compare differences between forests. These analyses showed no significant differences in timber yield between soil or physiographic units in Bennachie forest; in Kirkhill forest gley soils provided significantly greater yields than podzolic soils, with peaty gleys having significantly greater yields than other soil groups. In both forests, for virtually all chemical variables measured there is a significant increase from podzolic to gley soils, and from hill summits to low lying facets. Nutrient levels in the organic layer from profiles in Bennachie forest are about twice those recorded in Kirkhill. Correlation analysis was performed for all plots taken together and then on plot data subdivided into soil and physiographic units for each of Kirkhill and Bennachie forests. In Bennachie the site factors most consistently correlated with tree growth indices were acetic acid extract- able K and Mg in the organic sample and acetic extractable K in the mineral sample; all were negatively correlated. In Kirkhill, eastern aspects, depth of organic matter and total phosphate in both the organic and mineral sample were positively correlated with tree growth. Depth of rooting was negatively correlated. In general the subdivision of each forest into soil and physiographic units greatly reduces the number of significant correlations, particularly in gleys and peaty gleys and in low lying physiographic units. This reduction in correlations arises from the reduced variation within these units and also from the fewer degrees of freedom. From the factor analysis of Bennachie forest twelve factors were extracted, accounting for 80% of total variance. The first factor accounted for all the tree growth indices as well as acetic extractable K in the organic sample. In Kirkhill ten factors were extracted and accounted for 78% of the total variation. Again, the first factor contained all the tree growth indices along with eastern aspect, depth to mottling and rooting depth. The factor analysis thus summarizes the trends shown by the correlation analyses of close relationships between tree growth and soil chemical variables at Bennachie, whereas topographic and soil physical variables are more closely related to growth in Kirkhill. Abstract not available.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.464101  DOI: Not available
Share: