def extract_species_distributions(outfile):
"""
Extracts species distribution data from TNC catchment data
and writes to presence-absence matrix CSV
"""
# imports
import sys, arcpy, csv, numpy
# input files
hydroCodePath = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\Lamprey Cap Tradeoffs\analyses' # hydrogaphy code path
hydroMDB = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\Lamprey Cap Tradeoffs\raw_data\GL_pruned_hydrography.mdb'
sppData = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\Lamprey Cap Tradeoffs\raw_data\Catchments_with_FishPresence_Updated_4-25-2016\catchments_w_fish_presence_4-25-2016.shp'
# other parameters
cidField = 'HydroID'
speciesField = 'Species'
# import helper functions
cPath = sys.path
sys.path = [hydroCodePath] + cPath
from hydrography.hydrography import Hydrography
from hydrography.load_data import load_hydro_mdb
# load hydrography data
hydrodata = load_hydro_mdb(hydroMDB)
hydrography = Hydrography(hydrodata)
# load species trib data
spp2tid = {}
cid2tid = dict((c.id, c.tributary.id) for c in hydrography.get_catchments())
tids = sorted(set(cid2tid.values()))
ntid = len(tids)
tid2ind = dict((tids[i], i) for i in xrange(ntid))
for cid, species in arcpy.da.SearchCursor(sppData, [cidField, speciesField]):
tid = cid2tid.get(cid, None)
if tid is None: continue
species = species.strip()
if species == '': continue
if species not in spp2tid: spp2tid[species] = numpy.zeros(len(tids), dtype=numpy.uint8)
spp2tid[species][tid2ind[tid]] = 1
# write species data to output csv
species = sorted(spp2tid.keys())
writer = csv.writer(open(outFile, 'wb'))
header = [''] + tids
writer.writerow(header)
for s in species:
row = [s] + list(spp2tid[s])
writer.writerow(row)
return outFile
def main():
# ~~ IMPORTS ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #
from make_gdx import make_gdx, read_gms, load_data, prune_barriers
from hydrography.hydrography import Hydrography
from hydrography.load_data import load_hydro_mdb
import os
# ~~ DEFINE INPUTS ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #
# input files
barriers = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\Lamprey Cap Tradeoffs\raw_data\GL_pruned_hydrography.mdb\barriers'
tableCSV = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\code\benchmarks\data\table.csv'
defFile = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\code\benchmarks\data\definitions.csv'
gmsFile = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\code\benchmarks\scripts\Habitat_Opt.gms'
outFolder = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\code\benchmarks\data\gdxs'
hydroMDB = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\Lamprey Cap Tradeoffs\raw_data\GL_pruned_hydrography.mdb'
sppData = r'C:\Users\milt\Dropbox\UW Madison Post Doc\Lamprey Control\Lamprey Cap Tradeoffs\raw_data\Catchments_with_FishPresence_Updated_4-25-2016\catchments_w_fish_presence_4-25-2016.shp'
# other parameters
bidColumn = 'BID'
dsidColumn = 'BID_DS'
# get hydrography to map from bids to species presence
hydrodata = load_hydro_mdb(hydroMDB)
hydrography = Hydrography(hydrodata)
# assess which species might use which sections above which barriers
# (i.e. if that species is found in the river network of the barrier)
bid2spp = map_species_to_barriers(hydrography, sppData)
del hydrography, hydrodata
def spp_hab(sid, bid, hab):
if sid in bid2spp[bid]: return hab
else: return 0.
fish1_hab = lambda b,h: spp_hab('Esox lucius', b, h)
fish2_hab = lambda b,h: spp_hab('Moxostoma anisurum', b, h)
fish3_hab = lambda b,h: spp_hab('Acipenser fulvescens', b, h)
lamp_hab = lambda b,h: spp_hab('Petromyzon marinus', b, h)
# mapping from hydrography database fields to output fields for gams model
fields = {
'bid': ([bidColumn], lambda s: long(float(s))),
'dsid': ([dsidColumn], lambda s: long(float(s))),
'hab_fish1': ([bidColumn, 'HAB_UP'], fish1_hab),
'hab_fish2': ([bidColumn, 'HAB_UP'], fish2_hab),
'hab_fish3': ([bidColumn, 'HAB_UP'], fish3_hab),
'hab_lamprey': ([bidColumn, 'HAB_UP'], lamp_hab),
'pass_fish1': (['PASS04'], lambda s: s), # swimming guilds from Allison's paper
'pass_fish2': (['PASS07'], lambda s: s),
'pass_fish3': (['PASS10'], lambda s: s),
'pass_lamprey': (['PASS07'], lambda s: s),
'cost': (['COST'], lambda s: s),
'is_root': ([dsidColumn], lambda s: {-1: 1}.get(int(float(s)), 0)),
'candidate': ([], lambda: 1),
'passchange_fish1': (['PASS04'], lambda s: 1-float(s)),
'passchange_fish2': (['PASS07'], lambda s: 1-float(s)),
'passchange_fish3': (['PASS10'], lambda s: 1-float(s)),
'passchange_lamprey': (['PASS07'], lambda s: 1-float(s)),
'habchange_lamprey':([bidColumn, 'HAB_UP'], lambda b,h: -0.95*lamp_hab(b,h))
}
# ~~ CREATE GDX ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #
# make table.csv
make_table(barriers, fields, tableCSV)
# load the gams parameter definitions from the gams model file
gamsParameters = read_gms(gmsFile)
# load the data from tables and definitions file
data = load_data(tableCSV, defFile, gamsParameters)
# make the gdx's for every model run
print '\n'.join(make_gdx(
data, outFolder, parameters=gamsParameters, zip=False
))
