Premium
Predicting Norway Spruce Growth From Soil and Topographic Properties in New York
Author(s) -
Jokela E. J.,
White E. H.,
Berglund J. V.
Publication year - 1988
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1988.03615995005200030038x
Subject(s) - karst , productivity , soil water , regression analysis , regression , vegetation (pathology) , linear discriminant analysis , picea abies , environmental science , soil science , hydrology (agriculture) , linear regression , geography , mathematics , statistics , ecology , geology , biology , archaeology , medicine , geotechnical engineering , pathology , economics , macroeconomics
Soil‐site productivity relationships for Norway spruce ( Picea abies (L.) Karst.) were studied on a single, but widespread soil catena (Bath‐Lordstown‐Mardin‐Volusia; ∼1.1 million ha) in central New York. Soils and vegetation data collected from 37 unthinned plantations (45‐55‐yr old) were used to develop multiple regression soilsite predictor equations and discriminant classification functions. Efforts were made through sampling procedures to reduce extraneous sources of variation inherent in many previous soil‐site studies (e.g., parent materials, geographic area, stand density, stand age, cultural practices, and seed sources). Results of regression analyses showed that three to seven independent variables could explain 53 to 82% of the variation in mean annual volume increment (MAI). Stratifying the data set by a common planting density (2990 stems ha −1 ) generally improved the precision of the regression equations. Independent tests of the soil‐site equations across diverse drainage conditions showed a significant correlation between predicted and measured values of MAI ( r = 0.70). More accurate predictions resulted ( r = 0.80) when the equations were restricted to moderately well‐ and well‐drained soils only. Discriminant functions correctly reclassified (jackknifed analysis) 81% of the 37 stands into two MAI productivity groups. In general, results from the discriminant analyses supported interpretations made from the regression analyses. Soil properties, such as textural components, coarse fragments, pH, exchangeable cations, organic C, and cation exchange capacity (CEC), were most consistently correlated with Norway spruce volume production. Topographic properties made little or no independent contribution to the predictive models developed for growth. Results should aid yield predictions and site classification decisions for Norway spruce in New York.