Accounting for surveyor inconsistency and bias in estimation of tree density from presettlement land survey records

Abstract

Presettlement land survey records provide baseline data on forest characteristics prior to major European settlement, but questions regarding surveyor bias and methodological consistency have limited confidence in quantitative analyses of this important data source. We propose new correction factors, calculated from bearings, distances, and species of bearing trees, to account for the effects of (i) inconsistency in quadrant configuration, (ii) bearing angle bias, and (iii) species bias on forest density and species composition estimated from presettlement land survey records. Computer simulations confirmed accuracy in random and nonuniform density forests, with moderate bias in very clustered and dispersed forests. A case study of township and quarter-section corners surveyed by the Holland Land Company in western New York demonstrates the potential magnitude of errors caused by surveyor inconsistency/bias in estimation of density and relative species frequency. The influence of nonuniform density, clustering, and dispersal on plotless density estimators remains an important obstacle to quantitative analysis of Presettlement land survey records. However, by accounting for uncertainties regarding surveyor methodology, the proposed correction factors add confidence to conclusions made regarding presettlement forest structure and composition.