span.graph: Simulate landscape dynamics over a number of time steps
Description
This function gets an initial landscape graph and gradually applies changes. For a good review and classification of such changes see Bogaert et al. (2004) (not all described changes have been applied here). Future versions of the package should include other methods to change the landscape.
'hab' percentage of the number of patches to eliminate.
'dincr' minimal distance (between centroids of patches) increase over the simulation (in meters).
'darea' percentage of increase/decreas
par2
Parameter specifying details for the options in par1: percentage of patches do delete (if par1='hab'); distance, in meters (if par1='dincr'); percentage of increase/decrease (increase with negative sign) of the mean area of patches (if par1='darea'); perc
par3
Additional parameter specifying details for the options in par1: percentage of destroyed patches (if par1='stoc'); 'southerndness' of destroyed patches (if par1='ncsd'); aggregation of destruction (if par1='aggr'). Minimum area for patch deletion, in hect
par4
Percentage of created patches (if par1='ncsd').
par5
Percentage of destroyed patches (if par1='ncsd').
Value
Returns a list of data frames with the nodes characteristics of a given number of landscapes that suffer a specified change. The fields of these data frames are the same as those from the nodes characteristics resulting from rland.graph.
References
Bogaert, J., Ceulemans, R., & Salvador-Van Eysenrode, D. (2004). Decision tree algorithm for detection of spatial processes in landscape transformation. Environmental Management, 33(1): 62-73.
data(rland)
#Simulating a decrease of 5\% in the patches area through 100 time steps:span1 <- span.graph(rl=rland, span=100, par1="darea", par2=5, par3=NULL, par4=NULL, par5=NULL)