Developing distance-dependent growth models from irregularly measured sample plot data - A case for Larix olgensis in Northeast China

The need for distance-dependent individual-tree models for diameter increment and survival is increasing due to the increased use of individual-tree detection in forest inventories based on remote sensing data. Distance-dependent models can be used to analyze alternative thinning schemes, planting patterns and species mixing. The present study developed a methodology to develop distance-dependent models from data where the measurement interval of the sample plots is irregular. The method employs optimization to find the parameters of a simulation software in such a way that the simulated diameter distribution at the end of the measurement interval matches with the measured distribution. The methodology was applied to Larix olgensis A. Henry plantations of the Heilongjiang province of northeastern China. The distance-dependent models were fitted using two alternative competition indices adopted from earlier literature. The optimal way of computing the competition indices was solved simultaneously with the coefficients of model predictors. Distance-independent models were also fitted, and they provided a reference for assessing the performance of the distance-dependent models. The results indicated that the distance-dependent models did not provide better estimates on the periodical change of stand basal area or number of trees per hectare, compared to distance-independent models. However, the distance-dependent models are still useful as they enable model-based analyses of spatial questions.

Automated summary

This study developed a methodology to build distance-dependent models from irregularly sampled data, applied in Larix olgensis plantations in northeastern China. The distance-dependent models did not provide better estimates compared to distance-independent models, but are still useful for model-based spatial analysis.