Description Usage Arguments Details Value Author(s) References Examples
A stratified survey design that selects a set of sampling localities from a universe of available sites using an allocation procedure in which environmental and geographical distances are assumed to be surrogates for diversity variations. Environmental and geographical distances are combined prior to the calculations.
Selects the sampling points with a set of iterative rules. First step: Maximizises both the amount of spatio-environmental coverage using a p-median allocation procedure. Next steps: uses a set of rules or conditions defined by the user. Conditions should be related to the prioritization in the selection procedure (for example: conservation status or distance to roads). For each rule a vector of values and the type of criteria should be defined (see definition of criteria and conditions).
1 2 |
mdist |
A dissimilarity matrix coming usually from the combination of a matrix of environmental distances and a matrix of geographic distances. The maximum value of the matrices need to be approximately equal, so if one of the matrices has very high values, usually the geographic distances matrix, normalization is recommended to restrict the variation within 0 and 1. Once the matrices are normalized multiplying the two matrices is enough to get an even representation of selected sample sites. Missing values are not allowed. |
vini |
A binary vector (0,1) with the same length as the number of columns in
If |
vtarget |
A binary vector where values are equal to "1" are the sample sites fulfilling a set of minimum conditions defined by the user. For example, locations with good conservation status or an area above a threshold level. The length of the vector has to be the same as the number of columns in
|
criteria |
A vector of character strings specifying the criteria that will be applied
to the conditions vector. Currently available options are
For example, if the first condition (the first row in the conditions matrix)
is conservation status in a quantitative scale from 1 to 6 and we are
interested in well conserved sites we should chose The first criterion in the vector |
sdint |
A vector of numerical values with length equal to the number of criteria. By default the function defines a vector with all values equal to "1". So
one standard deviation will added or subtracted when applying the
continuous criteria ( To change the number of standard deviations added (in the case of
The value for ordinal criteria in the
Note that the first "1" in the vector applies to the ordinal criteria. See Example 2 in the examples Section for further explanations. Changes in the values of |
conditions |
A matrix with the same number of columns as Rules or conditions are defined by the user. A typical example is the level of conservation of each site. |
iter |
Number of iterations. Never bigger than the number of available sites, which are all the sites
minus the previously well sampled sites ( Note that the procedure selects one site per iteration, so the number of iterations and the number of selected points will be the same. |
In each iteration, the procedure first selects a set of sample sites
minimizing the total distance between the non selected sites (p-median
procedure), and then, in the next steps, criteria defined by the user in
conditions
are used iteratively to restrict the number of selected
sites. If more than one point remain selected after applying all the
conditions, the function selects one at random. Therefore, in each step just
one sample site is selected. The next iteration will start with all the
points minus the selected point in the previous iteration and so on.
The function prints the criterion used to select the sampling point in each
iteration. By default the names of the criteria arecrit1
,
crit2
, etc.
crit1
represents the first criterion included in the criteria
vector, crit2
the second and so on.
The function also returns a list including two matrices: pmmatrix
,
selmatrix
(see below).
selmatrix |
|
pmmatrix |
The value of each cell is the distance between each site to the closest previously selected site. Note that selected sites in each step have "0" value in this matrix. Adding the values of the rows (see function |
Nagore Garcia Medina & Bernardo Garcia Carreras
Church, R.L. & Sorensen, P. (1994) Integrating Normative location models into GIS: problems and prospects with the p-median model. Technical Report, NGCIA.
Church, R.L. (2002) Geographical information systems and location science. Computers and Operation Research 29: 541-562.
Faith, D.P. & Walker, P.A. (1996) Environmental diversity: on the best possible use of surrogate data for assessing the relative biodiversity of sets of areas. Biodiversity and Conservation 5: 399-415.
Hortal, J., Araujo, M.B. & Lobo, J.M. (2009) Testing the effectiveness of discrete and continuous environmental diversity as a surrogate for species diversity. Ecological Indicators 9: 139-149.
Hortal, J. & Lobo, J.M. (2005) An ED-based protocol for the optimal sampling of diversity. Biodiversity and Conservation 14: 2913-2947.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 | # load the environmental and spatial matrices
data(env)
data(geogr)
# multiply environmental and geographic matrices
mdist <- env * geogr
# load a matrix containing the conditions, the initial vector and the target
# points
data(conditions)
# define vector of initial points
data(vini)
# define the criteria to apply to the conditions: maximise conservation
# (qualitative variable), maximise area of the site (qualitative variable)
# and minimise distance to roads (quantitative variable)
data(criteria)
result <- alloc(mdist=mdist, vini=vini, criteria=criteria,
conditions=conditions, iter=20)
uncov <- rowSums(result$pmmatrix)
plot(1:length(uncov), uncov)
# Example 2
# With the same data as Example 1, but with a change in the number of
# standard deviations added in the first criteria, note that the first
# condition is slope of the species area curve, and the criteria to apply is
# \code{rangemax}, in this case the range of selection for the condition will
# be = max-0.1*sd
result <- alloc(mdist=mdist, vini=vini, criteria=criteria, sdint=c(0.1,1,1),
conditions=conditions, iter=20)
uncov <- rowSums(result$pmmatrix)
plot(1:length(uncov), uncov)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.