The ability to inventory, monitor and assess natural resources is often constrained by limited resources.  The efficiency and cost-effectiveness of monitoring and assessment can be increased by optimizing the number and arrangement of samples collected. This requires the planner to supply such metrics as:  the magnitude of the difference to be detected (minimum detectable difference, MDD), the variability of the property (measured as variance of the population within the project inference space), and the desired level of precision, acceptable error rates, and correlation between samples being compared (a, b and rho respectively). These needs require careful consideration of the inference space of the project.  For this study, we use ecological sites and status (deviation of vegetation from a reference state) to define inference space and the Multi-Scale Sampling Size Requirements Evaluation tool, MSSRET, a web- or spreadsheet- based tool developed by NRCS and ARS, to evaluate sampling requirements across those inference spaces.  Vegetation metrics and 18 soil samples were collected on 1 ha plots. The magnitude and scale of variance varied with property, depth, and ecological site. For carbon stocks (Mg C ha^-1), MDD was set to typical 5 year contract values for rangeland carbon offsets. In order to detect a change of 1.75 Mg C ha-1, MSSRET plot recommendations range from 96 plots for all ecological sites and conditions to 2 plots for the degraded condition of the Sandy ecological site when 10 samples are collected in each plot.  While limiting the plot size, as sometimes recommended in agronomic systems, would likely reduce the soil variance and the sampling requirements, this would reduce the inference space and area to which the results could be applied.  The patchy nature of rangeland vegetation and soil carbon responses to management requires either large plots or multiple sets of plots stratified by landscape and vegetation features to capture variability in a way that can be reasonably extrapolated over larger landscapes.