Sample design: Difference between revisions
Jump to navigation
Jump to search
| Line 16: | Line 16: | ||
The stratification is done in each domain separately. The stands (their productive forest land areas) of a certain domain are related to one of the several user-defined strata. In addition to total area, the number of stands and the area of the largest stand can be viewed in each stratum. The stratum-matrix should first be arranged according to current forest state (by number of classes and class width) and, if desirable, defined by other variables than volume and age. | The stratification is done in each domain separately. The stands (their productive forest land areas) of a certain domain are related to one of the several user-defined strata. In addition to total area, the number of stands and the area of the largest stand can be viewed in each stratum. The stratum-matrix should first be arranged according to current forest state (by number of classes and class width) and, if desirable, defined by other variables than volume and age. | ||
=== | ===Number of Sample Stands=== | ||
This is called the sample size-definition part. First, a defined number of sample stands will be allocated to each stratum (e.g., proportional to the volume). You could and probably should re-distribute the sample stands over the different strata, with a higher sampling intensity in strata with more valuable forests and where forest management decisions are supposed to be more difficult. The term and parameter "Representative Area", with a value calculated as ''Productive forest land area in current stratum / Sample size (number of sampled stands) in current stratum'', might be found a bit confusing. ''A scale factor with a unit (hectares)? With the same value for all sampled stands in a stratum (despite they probably will have different areas)?'' Don't worry, as long as the sampling of stands is done PPS, and size is the productive area (in hectares), it will all turn out alright. Try to follow this: an estimation of, e.g., the total biomass (in m<sup>3</sup>) in a stratum is derived using a Horvitz-Thompson (HT) estimator with inclusion probabilities (''�?''<sub>''i''</sub>) defined as ''(Area of the i:th sample stand * Number of sample stands in current stratum) / Area of current stratum''. By dividing this estimation by the total area of the stratum an average value (here in m<sup>3</sup>/ha) is derived. Finally, the total volume of biomass in current stratum that is represented by the sample is derived by multiplying the average value-estimate by ''Representative area''. The only time this procedure is not followed is when a single stand show an area equal to or greater than the representative area - such stand will become a sample stand and represent its own area (nothing more, nothing less). | This is called the sample size-definition part. First, a defined number of sample stands will be allocated to each stratum (e.g., proportional to the volume). You could and probably should re-distribute the sample stands over the different strata, with a higher sampling intensity in strata with more valuable forests and where forest management decisions are supposed to be more difficult. The term and parameter "Representative Area", with a value calculated as ''Productive forest land area in current stratum / Sample size (number of sampled stands) in current stratum'', might be found a bit confusing. ''A scale factor with a unit (hectares)? With the same value for all sampled stands in a stratum (despite they probably will have different areas)?'' Don't worry, as long as the sampling of stands is done PPS, and size is the productive area (in hectares), it will all turn out alright. Try to follow this: an estimation of, e.g., the total biomass (in m<sup>3</sup>) in a stratum is derived using a Horvitz-Thompson (HT) estimator with inclusion probabilities (''�?''<sub>''i''</sub>) defined as ''(Area of the i:th sample stand * Number of sample stands in current stratum) / Area of current stratum''. By dividing this estimation by the total area of the stratum an average value (here in m<sup>3</sup>/ha) is derived. Finally, the total volume of biomass in current stratum that is represented by the sample is derived by multiplying the average value-estimate by ''Representative area''. The only time this procedure is not followed is when a single stand show an area equal to or greater than the representative area - such stand will become a sample stand and represent its own area (nothing more, nothing less). | ||
''NB: There should be at least three sample stands in each stratum. Moreover, aviod extreme differences in the representative area of different strata.'' | ''NB: There should be at least three sample stands in each stratum. Moreover, aviod extreme differences in the representative area of different strata.'' | ||
===Plot Intensity=== | ===Plot Intensity=== | ||