Five population‐specific response functions were developed from quadratic models for 110 populations of Pinus sylvestris growing at 47 planting sites in Eurasia and North America. The functions predict 13 year height from climate: degree‐days > 5 °C; mean annual temperature; degree‐days < 0 °C; summer‐winter temperature differential; and a moisture index, the ratio of degree‐days > 5 °C to mean annual precipitation. Validation of the response functions with two sets of independent data produced for all functions statistically significant simple correlations with coefficients as high as 0.81 between actual and predicted heights. The response functions described the widely different growth potentials typical of natural populations and demonstrated that these growth potentials have different climatic optima. Populations nonetheless tend to inhabit climates colder than their optima, with the disparity between the optimal and inhabited climates becoming greater as the climate becomes more severe. When driven by a global warming scenario of the Hadley Center, the functions described short‐term physiologic and long‐term evolutionary effects that were geographically complex. The short‐term effects should be negative in the warmest climates but strongly positive in the coldest. Long‐term effects eventually should ameliorate the negative short‐term impacts, enhance the positive, and in time, substantially increase productivity throughout most of the contemporary pine forests of Eurasia. Realizing the long‐term gains will require redistribution of genotypes across the landscape, a process that should take up to 13 generations and therefore many years.

This reprinting of the User's Guide to the Prognosis Model describes the Prognosis Model as released in September, 1981 (Version 4.0). Although we will soon release version 5.0, most of the material in this guide will remain applicable to the new version. There will be, however, modifications in the small tree growth models and in the crown-dubbing and crown-changing procedures that improve model behavior. These modifications will necessitate revisions of pages 52, 65-67, and 77-80.These revisions, and descriptions of new features, are contained in a supplement to this guide that will be released with the new version. The new features include:-A regeneration establishment component;-SHRUB and COVER extensions;-An event monitor for dynamic activity scheduling;-A classification algorithm used to shorten the tree record list by combining like records;-Expansion of management options.We have endeavored to make changes in such a way that the procedures for using version 4.0 will operate the same way in version 5.0.

Version 4.0 of the Prognosis Model was released in September 1981. Since then, a regeneration establishment model has been completed and small-tree increment models have been greatly refined. The COVER model has also been added to predict shrub development and total canopy cover. Thus, the representation of the vegetative component of the stand is basically complete and the Stand Prognosis Model can be linked more readily to models for nontimber resources.New management options have been added to the system, and an Event Monitor increases the flexibility for scheduling management activities. A compression or classification algorithm enhances program efficiency by combining tree records that are similar with regard to attributes that influence growth predictions. Finally, there have been numerous improvements in the biological models.This report is a supplement to the user's guide for the Stand Prognosis Model (Wykoff and others 1982). Options that were available in version 4.0 may still be invoked in the manner described in the user's guide. New options, new models, and modifications to existing models are herein described as incorporated in version 5.0.

This paper proposes a method whereby height–diameter regression from an inventory can be incorporated into a height imputation algorithm. Point-level subsampling is often employed in forest inventory for efficiency. Some trees will be measured for diameter and species, while others will be measured for height and 10-year increment. Predictions of these missing measures would be useful for estimating volume and growth, respectively, so they are often imputed. We present and compare three imputation strategies: using a published model, using a localized version of a published model, and using best linear unbiased predictions from a mixed-effects model. The bases of our comparison are four-fold: minimum fitted root mean squared error and minimum predicted root mean squared error under a 2000-fold cross-validation for tree-level height and volume imputations. In each case the mixed-effects model proved superior. This result implies that substantial environmental variation existed in the height–diameter relationship for our data and that its representation in the model by means of random effects was profitable.

No abstract