Comparison of Variance Estimators for Systematic Environmental Sample Surveys: Considerations for Post-Stratified Estimation

The estimation of the sampling variance point estimators under two-dimensional systematic designs remains a challenge, and several alternative have been proposed in past few decades. In this work, we compared six Horvitz-Thompson (HT) post-stratification (PS) regimes. We subsampled multitude species-specific forest attributes from large, spatially balanced national inventory to compare estimators. A estimator that assumes simple random design exhibited positive relative bias both HT PS regimes ranging between 1.23 1.88 1.11 1.78 for PS, respectively. Alternative reduced with biases 1.01 1.66 0.90 1.64 generally obtained improved efficiencies efficiency values 0.68 1.28 1.39, identified two as promising alternatives provide clear improvements over wide variety