Calculate the difference in the landscape-level metric totals between a baseline landscape and one or more scenario landscapes
Value
tibble containing fields SCENARIO_VALUE, BASELINE_VALUE,
net_change, and optionally SCENARIO_SE, BASELINE_SE, net_change_se,
U, lcl, ucl, and z; see Details
Details
This function expects dat to contain the following fields:
scenario: a character field used to identify the name of the landscape being examined; one of these must be called'baseline'and all others will be treated as alternate scenarios for comparison with the baselineeither
areaorSCORE_TOTAL: numeric fields representing the total landscape-level values for each metric, such as those produced bysum_landcover()orsum_metrics().
The scores for each scenario are aligned with corresponding scores for the
baseline landscape based on any other common fields (e.g. ZONE,
METRIC_CATEGORY) and the net difference is calculated as the scenario
score minus the baseline score. The function returns the original baseline
and scenario total landscape scores for each metric and scenario (renamed as
BASELINE and SCENARIO), along with net_change.
If SCORE_TOTAL_SE is also provided in dat, representing the uncertainty
in the SCORE_TOTAL, the uncertainty in the difference (net_change_se) is
also calculated as: sqrt(BASELINE_SE^2 + SCENARIO_SE^2) where
BASELINE_SE and SCENARIO_SE represent the SCORE_TOTAL_SE for the
baseline and scenario landscapes, respectively. In addition, the coverage
factor k is used to estimate expanded uncertainty (U), or the interval
within which a large fraction of the distribution of values could be
reasonably expected. The appropriate value for k depends on the level of
confidence required, the number of observations on which the uncertainty is
based, and any knowledge of the underlying distributions of the estimates.
Where the distributions concerned are normal, and for most purposes, a value
of 2 is recommended to give an interval containing approximately 95% of the
distribution of values. In this case, the function also returns the original
uncertainty estimates for the baseline and scenario (renamed as
BASELINE_SE and SCENARIO_SE), net_change_se, the expanded uncertainty
estimate U and corresponding lcl and ucl, as well as a z score equal
to abs(net_change/net_change_se).