The US Public Land Survey (PLS) data from the US General Land Office are one of the best records of the pre-European settlement vegetation. These records have been used to generate pre-settlement vegetation maps in many states. The purpose of this study was to examine the spatial pattern of tree species and their scaling relationships with environmental variables across environments in the state of Wisconsin by using PLS records.  We used point pattern analyses, wavelet and other spatial analyses to quantify the hierarchical spatial pattern of tree species at multiple scales. The detected spatial patterns were then examined with soil variables (extracted from SSURGO) and other environmental variables. 
Results/Conclusions

The results provide a framework of how environmental variables influence spatial distribution of tree species at multiple scales under minimum human disturbances.  In the north, physical variables predominate in importance, while in the south there is less congruence with physical variables, apparently due to higher past fire disturbances. Across the state, species composition is strongly related to both temperature and precipitation gradients, within which finer scale patterns are nested.