Nature as a model for large-scale planting design – Least-difference method

An objective and systematic method for large-scale planting design was previously published. It involved multiple regression equations relating species abundance to topo-edaphic variables in a native (model) plant community, calculated from survey data collected by forestry departments. The method is appropriate for the number of sample plots currently surveyed in plant communities. With the imminent availability of a large volume of data through remote sensing technology, another method is presented in this article that emulates nature more closely and is not based on a mathematical model. It involves finding sample plots in the model community with topo-edaphic variables similar to those of the grid cells on the project site and assigning their species composition to the grid cells. Similarities of corresponding pairs of variables are decided using comparison tolerances for the variables, calculated based on their relative importance in species distribution in the model community.