Soil variability on steep, skeletal forested slopes of the Umpqua National Forest, Oregon

Graduation date: 1983This study was undertaken to determine both the amount and the\ud spatial variability of sand, silt, clay, and six fractions of coarse\ud fragments in soils representing harsh sites for reforestation. Particle size data were collected because of their direct relationship to\ud water holding capacity and the plantability and survivability of tree\ud seedlings. At each of four sites, samples of both surface and sub-surface soil were taken at 15 meter intervals on a 10 x 10 grid.\ud Particle size distributions have both binomial and continuous\ud attributes. The binomial character reflects the presence or absence\ud of a given particle size fraction, and is most directly related to the\ud presence of bedrock outcrops, rock streams, and talus deposits in the\ud landscape. Absence of a fraction creates a zero in the data set. Zero\ud values skew many of the distributions and distort calculated measures\ud of central tendency and variation. Exclusion of zeroes normalizes most\ud distributions and reduces the variability significantly. The two largest coarse fragment fractions remain positively skewed. For them, the\ud mode is a better measure of central tendency than the mean. Proper evaluation of soil variability thus requires determination of the binomial\ud frequency of occurrence of each particle size fraction, followed by\ud calculation of the mode and the variance of all the non-zero values. Correlations between surface soil properties and subsoil properties\ud ranged from .05 to .71. Most were between .35 and .55. Subsoil conditions, therefore, cannot be predicted satisfactorily from a knowledge of\ud surface soil conditions.\ud The U.S. Forest Service has proposed four criteria for evaluating\ud physical site plantability. All are based on the amount and size of\ud coarse fragments present. Meeting any one renders a soil unsuited for\ud reforestation. Data from this study indicate that the single criterion\ud of 80% or more by volume total coarse fragments is the only one necessary to determine plantability. It is sufficiently comprehensive to include all possible unplantable soils. Field sieving to determine total\ud coarse fragments is very slow, but plantability can be quite accurately\ud determined in the field by sieving through a 9.5 mm. sieve and weighing\ud the coarse fragments retained. This procedure requires a minimum of\ud equipment and only 15 minutes of time, and it corresponds to total\ud coarse fragment content with an error rate of less than 5%.\ud The binomial character of particle size distributions suggests a\ud two-stage sampling scheme to evaluate site plantability. Stage one\ud determines the frequency of rock outcrop and talus presence by observing at 30 meter intervals on a pace-compass grid. Stage two involves\ud soil sampling and/or field analysis at points within the non-rock out\ud crop area. Less than 10 samples would be required to estimate the\ud true mean content of most fractions within ± 5% at 90% confidence. A\ud larger sample size is needed for the two largest coarse fractions