Skip to content
Login has been disabled on EnviDat Legacy. Please log in via https://envidat.ch first, then refresh this page.

Changes

View changes from to


On May 15, 2023 at 9:02:12 AM UTC, Gravatar Klaus Ecker:
  • Updated description of R script and input data for "ALL-EMA sampling design" from

    License: GPL-v2 The R script illustrates an avanced sampling approach for monitoring biodiversity on agricultural land by combining multiple objectives and integrating environmental and geographic space for optimal estimation efficiency. In doing so, the R script demonstrates the first-stage selection of squares (km2) in the ALL-EMA sampling design using modern sampling techniques such as unequal probability sampling with fixed sample size, balanced sampling, stratified balancing and geographic spreading. Sampling is done with unequal probabilities and weights defined by power allocation to give equal weight to extrapolations to the total agricultural area of Switzerland and two stratifications of predefined interest (regions and agricultural production zones). Calibration is used to limit the distribution of the sampling weights. The sample sizes are almost fixed within the strata and evenly distibuted across the years of a temporal rotation plan, which is favourable for the organisation of the field survey. Sampling also ensures an optimal (annual) distribution across geographic space, including altitude. Despite the complexity of the sampling, estimation based on probability theory is straightforward. Ecker, K., Meier, E. & Tillé, Y. review. Integrating spatial and ecological information into comprehensive biodiversity monitoring on agricultural land. Environmental Monitoring and Assessment.
    to
    License: GPL-v2 The R script presents an advanced sampling approach for monitoring biodiversity on agricultural land by combining multiple objectives and integrating environmental and geographic space. The example demonstrates the first-stage selection of squares (km2) in the ALL-EMA sampling design using modern sampling techniques such as unequal probability sampling with fixed sample size, balanced sampling, stratified balancing and geographic spreading. Sampling is done with unequal probabilities and weights defined by power allocation to give equal weight to extrapolations to the total agricultural area of Switzerland and two stratifications of predefined interest (regions and agricultural production zones). Calibration is used to limit the distribution of the sampling weights. The sample sizes are almost fixed within the strata and evenly distributed across the years of a temporal rotation plan, which is favourable for the organisation of the field survey. Sampling also ensures an optimal (annual) distribution across geographic space, including altitude. Despite the complexity of the sampling, estimation based on probability theory is straightforward. Ecker, K., Meier, E. & Tillé, Y. review. Integrating spatial and ecological information into comprehensive biodiversity monitoring on agricultural land. Environmental Monitoring and Assessment.