Spatial Autocorrelation in Growth of Undisturbed Natural Pine Stands Across Georgia

Moran's I statistic measures the spatial autocorrelation i a random variable measured at discrete locations in space. Permutation procedures test the null hypothesis that the observed Moran's I value is no greater than that expected by chance. The spatial autocorrelation of gross basal area increment is analyzed for undisturbed, naturally regenerated stands in three Georgia forest ypes: 1oblolly, shortleaf, and slash pine. The analysis uses 0.4-ha permanent sample plots from a forest inventory that included two remeasurement intervals (1961-1972 and 1972-1982). We present a new statistic for exploratory spatial analyses, and this statistic revealed an anomalous cluster of unusually slowgrowing shortleaf pine plots occurred in the mountains 100 km north of Atlanta. A regression model was used to predict gross basal area increment as a function of variables that describe local stand conditions, and no significant spatial autocorrelations existed in the regression residuals. This result suggests that the anomalous cluster of slow-growing plots can be explained by the spatial distribution of local stand conditions rather than spatial patterns of other possible causes such as air pollution, although alternative interpretations are possible. FOR. SCI. 40(2):314-328. B ECHTOLD ET AL. (1991) REPORTED ECR ASES IN ADJU TED MEAN GROWTH of naturally regenerated undisturbed pine plots in the State of Georgia (150,000 km 2) using forest inventory plots. Between the growth intervals 1961-1972 and 1972-1982, the decreases in pine basal area increment per unit area were 19% for 1oblolly pine and 28% for slash and shortleaf pine after adjustments for local stand conditions. They were unable to resolve the role of exogenous factors, such as climate or pollution, as possible causes of reduced growth. They suggest hat "several factors are contributing to the reductions, probably varying in importance with species, location, and time." However, Hyink (1991) suspects that growth rates computed for 1961-1972 are unusually high, and the observed growth between 1972-1982 represents a return to normal conditions. Lucier and Barnard (1992) conclude that the reduction in growth of certain classes of natural pine stands is "an expected consequence of historical land use patterns, increases in stand age and competition, and other non-pollutant factors," and an abnormal "regional decline of southern pines has not been demonstrated." We conjectured that spatial analyses might reveal nonrandom patterns in the distribution of growth across Georgia that would be consistent or inconsistent with one or more of the many competing hypotheses.