notes_system_models

# need to add priors, tell model what the system is doing. 
# water draw down does things with the available prey concentrations
# birds need x amount of food to start nesting, laying
# there is a mosaic of water bodies that dry up at different rates
# largest scale is cardinal directions, smallest is GIS locations 

#water levels affect prey fish and bird populations differently 
#in different time periods. 

# fish(t-5) water is HIGH,  
#     large fish effect prey 
# fish(t-4) water HIGH-MEDIUM generating topographic hiding locations in landscape
#     effect of large fish goes down, more prey fish 
#fish(t-3) water MEDIUM, more hiding paces for prey. 
#     small effect from large fish, highest amount of pry fish
#fish(t-2) water LOW-MEDIUM 
#     prey start becoming available for birds, predator switch on prey fish 
#fish(t-1) water LOW, prey becomes constantly available for bird predation   
#     high effect from birds on prey fish 
#fish(t-0) water LOW-LOW, water mass can't contain fish, less fish for birds
#     negative effects on bird population 

#birds(t-5) water is HIGH, 
#     to high water levels for most species to forage, available around shore
#birds(t-4) water HIGH-MEDIUM
#     more areas become available, bathymetry can allow for spots
#birds(t-3) water MEDIUM
#     more areas become available, bathymetry can allow for spots
#birds(t-2) water LOW-MEDIUM 
#     high availability in the landscape for prey availability 
#birds(t-1) water LOW
#     highest availability in the landscape for prey availability, high pressure on prey 
#birds(t-0) water LOW-LOW
#     not enough water to sustain prey population, high negative effect on birds 



=============================================================================
# subgr	
# A subgrouping factor variable specifying which element 
# in data represents the different time series. Defaults 
# to series, but note that models that use the hierarchical 
# correlations, where the subgr time series are measured in 
# each level of gr, should not include a series element in 
# data. Rather, this element will be created internally 
# based on the supplied variables for gr and subgr.
# 
# For example, if you are modelling temporal counts for a 
# group of species (labelled as species in data) across 
# three different geographical regions (labelled as region), 
# and you would like the residuals to be correlated within 
# regions, then you should specify gr = region and 
# subgr = species. Internally, mvgam() will create the 
# series element for the data using:
#   series = interaction(group, subgroup, drop = TRUE)

=============================================================================
#zero inflated NB - needed? - probably if we look at the regions