delr=swt_delr, delc=swt_delc, laycbd=0, top=swt_top, botm=swt_botm,
nper=nper, perlen=perlen, nstp=1, steady=False)
bas = mf.ModflowBas(ml, ibound=swt_ibound, strt=0.05)
lpf = mf.ModflowLpf(ml, hk=2.0, vka=2.0, ss=0.0, sy=0.0, laytyp=0, layavg=0)
oc = mf.ModflowOc88(ml, save_head_every=1, item2=[[0, 1, 0, 0]])
pcg = mf.ModflowPcg(ml)
ml.write_input()
# Create the basic MT3DMS model structure
mt = mt3.Mt3dms(modelname, 'nam_mt3dms', modflowmodel=ml,
model_ws=dirs[1]) # Coupled to modflow model 'mf'
adv = mt3.Mt3dAdv(mt, mixelm=-1, #-1 is TVD
percel=0.05,
nadvfd=0, #0 or 1 is upstream; 2 is central in space
#particle based methods
nplane=4,
mxpart=1e7,
itrack=2,
dceps=1e-4,
npl=16,
nph=16,
npmin=8,
npmax=256)
btn = mt3.Mt3dBtn(mt, icbund=1, prsity=ssz, sconc=sconc, ifmtcn=-1,
chkmas=False, nprobs=10, nprmas=10, dt0=0.0, ttsmult=1.2, ttsmax=100.0,
ncomp=1, nprs=nprs, timprs=timprs, mxstrn=1e8)
dsp = mt3.Mt3dDsp(mt, al=0., trpt=1., trpv=1., dmcoef=0.)
gcg = mt3.Mt3dGcg(mt, mxiter=1, iter1=50, isolve=3, cclose=1e-6, iprgcg=5)
ssm = mt3.Mt3dSsm(mt, stress_period_data=ssm_data)
mt.write_input()
# Create the SEAWAT model structure
mswtf = swt.Seawat(modelname, 'nam_swt', modflowmodel=ml, mt3dmodel=mt,
exe_name=swtexe_name, model_ws=dirs[1]) # Coupled to modflow model mf and mt3dms model mt
rclose=0.001,
relax=1.0,
nbpol=0)
ml.write_input()
'''Initialize MT3DMS packages'''
mt = mt3.Mt3dms(name,
'nam_mt3dms',
exe_name=swtexe_name,
modflowmodel=ml,
model_ws=dirs[0])
adv = mt3.Mt3dAdv(
mt,
mixelm=0,
percel=0.8,
nadvfd=1, #Particle based methods
nplane=0,
mxpart=250000,
dceps=1e-4,
npl=5,
nph=8,
npmin=1,
npmax=16)
btn = mt3.Mt3dBtn(
mt,
cinact=-100.,
icbund=grid_obj.ICBUND,
prsity=grid_obj.PEFF,
sconc=grid_obj.temp, #sconc2=grid_obj.salinity,
ifmtcn=-1,
chkmas=False,
nprobs=0,
nprmas=1,
def get_package(self, _mt):
content = self.merge()
return mt.Mt3dAdv(
_mt,
**content
)
def run_experiment(self, experiment):
'''
Method for running an instantiated model structure.
This method should always be implemented.
:param case: keyword arguments for running the model. The case is a
dict with the names of the uncertainties as key, and
the values to which to set these uncertainties.
'''
#NetLogo agent attributes to be passed to Python well objects
#when new wells are created in NetLogo
nl_read_sys_attribs = ['who', 'xcor', 'ycor']
nl_read_well_attribs = [
'who', 'xcor', 'ycor', 'IsCold', 'z0', 'FilterLength', 'T_inj', 'Q'
]
#NetLogo agent attributes to be updated by the Python objects after each period
nl_update_well_attribs = ['T_modflow', 'H_modflow']
nl_update_globals = ['ztop', 'Laquifer']
self.netlogo.command('setup')
for key, value in experiment.items():
if key in self.NetLogo_uncertainties:
try:
self.netlogo.command(self.command_format.format(
key, value))
except jpype.JavaException as e:
warning('Variable {0} throws exception: {}'.format(
(key, str(e))))
logging.debug(self.netlogo.report(str(key)))
if key in self.SEAWAT_uncertainties:
setattr(self, key, value)
#Set policy parameters if present
if self.policy:
for key, value in self.policy.items():
if (key in self.NetLogo_uncertainties and key != 'name'):
self.netlogo.command(self.command_format.format(
key, value))
elif key in self.SEAWAT_uncertainties:
setattr(self, key, value)
logging.info('Policy parameters set successfully')
#Update NetLogo globals from input parameters
for var in nl_update_globals:
self.netlogo.command(
self.command_format.format(var, getattr(self, var)))
#Run the NetLogo setup routine, creating the agents
#Create lists of Python objects based on the NetLogo agents
self.netlogo.command('init-agents')
sys_obj_list = update_runtime_objectlist(self.netlogo, [],
nl_read_sys_attribs,
breed='system',
objclass=PySystem)
well_obj_list, newgrid_flag = update_runtime_objectlist(
self.netlogo, [],
nl_read_well_attribs,
breed='well',
objclass=PyWell)
#Assign values for uncertain NetLogo parameters
logging.info('NetLogo parameters set successfully')
#self.netlogo.command('init-agents')
#Calculate geohydrological parameters linked to variable inputs
rho_b = self.rho_solid * (1 - self.PEFF)
kT_b = self.kT_s * (1 - self.PEFF) + self.kT_f * self.PEFF
dmcoef = kT_b / (self.PEFF * self.rho_f * self.Cp_f) * 24 * 3600
trpt = self.al * self.trp_mult
trpv = trpt
#Initialize PyGrid object
itype = mt3.Mt3dSsm.itype_dict()
grid_obj = PyGrid()
grid_obj.make_grid(well_obj_list,
dmin=self.dmin,
dmax=self.dmax,
dz=self.dz,
ztop=self.ztop,
zbot=self.zbot,
nstep=self.nstep,
grid_extents=self.grid_extents)
#Initial arrays for grid values (temperature, head) - for this case, assumes no groundwater flow
#and uniform temperature
grid_obj.ncol = len(grid_obj.XGR) - 1
grid_obj.delr = np.diff(grid_obj.XGR)
grid_obj.nrow = len(grid_obj.YGR) - 1
grid_obj.delc = -np.diff(grid_obj.YGR)
grid_obj.top = self.ztop * np.ones([grid_obj.nrow, grid_obj.ncol])
botm_range = np.arange(self.zbot, self.ztop, self.dz)[::-1]
botm_2d = np.ones([grid_obj.nrow, grid_obj.ncol])
grid_obj.botm = botm_2d * botm_range[:, None, None]
grid_obj.nlay = len(botm_range)
grid_obj.IBOUND, grid_obj.ICBUND = boundaries(
grid_obj) #Create grid boundaries
#Initial arrays for grid values (temperature, head)
init_grid = np.ones((grid_obj.nlay, grid_obj.nrow, grid_obj.ncol))
grid_obj.temp = 10. * init_grid
grid_obj.HK = self.HK * init_grid
grid_obj.VK = self.VK * init_grid
#Set initial heads according to groundwater flow (based on mfLab Utrecht model)
y_array = np.array([(grid_obj.YGR[:-1] - np.mean(grid_obj.YGR[:-1])) *
self.PEFF * -self.gwflow_y / 365 / self.HK])
y_tile = np.array([np.tile(y_array.T, (1, grid_obj.ncol))])
x_array = (grid_obj.XGR[:-1] - np.mean(
grid_obj.XGR[:-1])) * self.PEFF * -self.gwflow_x / 365 / self.HK
y_tile += x_array
grid_obj.head = np.tile(y_tile, (grid_obj.nlay, 1, 1))
#Set times at which to read SEAWAT output for each simulation period
timprs = np.array([self.perlen])
nprs = len(timprs)
logging.info('SEAWAT parameters set successfully')
#Iterate the coupled model
for period in range(self.run_length):
#Set up the text output from NetLogo
commands = []
self.fns = {}
for outcome in self.outcomes:
#if outcome.time:
name = outcome.name
fn = r'{0}{3}{1}{2}'.format(self._working_directory, name,
'.txt', os.sep)
self.fns[name] = fn
fn = '"{}"'.format(fn)
fn = fn.replace(os.sep, '/')
if self.netlogo.report('is-agentset? {}'.format(name)):
#If name is name of an agentset, we
#assume that we should count the total number of agents
nc = r'{2} {0} {3} {4} {1}'.format(fn, name, 'file-open',
'file-write', 'count')
else:
#It is not an agentset, so assume that it is
#a reporter / global variable
nc = r'{2} {0} {3} {1}'.format(fn, name, 'file-open',
'file-write')
commands.append(nc)
c_out = ' '.join(commands)
self.netlogo.command(c_out)
logging.info(' -- Simulating period {0} of {1}'.format(
period, self.run_length))
#Run the NetLogo model for one tick
self.netlogo.command('go')
logging.debug('NetLogo step completed')
#Create placeholder well list - required for MODFLOW WEL package if no wells active in NetLogo
well_LRCQ_list = {}
well_LRCQ_list[0] = [[0, 0, 0, 0]]
ssm_data = {}
ssm_data[0] = [[0, 0, 0, 0, itype['WEL']]]
#Check the well agents which are active in NetLogo, and update the Python objects if required
#The newgrid_flag indicates whether or not the grid should be recalculated to account for changes
#in the list of active wells
if well_obj_list:
well_obj_list, newgrid_flag = update_runtime_objectlist(
self.netlogo, well_obj_list, nl_read_well_attribs)
if well_obj_list and newgrid_flag:
#If the list of active wells has changed and if there are active wells, create a new grid object
newgrid_obj = PyGrid()
newgrid_obj.make_grid(well_obj_list,
dmin=self.dmin,
dmax=self.dmax,
dz=self.dz,
ztop=self.ztop,
zbot=self.zbot,
nstep=self.nstep,
grid_extents=self.grid_extents)
#Interpolate the temperature and head arrays to match the new grid
newgrid_obj.temp = grid_interpolate(grid_obj.temp[0, :, :],
grid_obj, newgrid_obj)
newgrid_obj.head = grid_interpolate(grid_obj.head[0, :, :],
grid_obj, newgrid_obj)
#Use the new simulation grid
grid_obj = newgrid_obj
logging.debug('Python update completed')
if well_obj_list:
for i in well_obj_list:
#Read well flows from NetLogo and locate each well in the simulation grid
i.Q = read_NetLogo_attrib(self.netlogo, 'Q', i.who)
i.calc_LRC(grid_obj)
#Create well and temperature lists following MODFLOW/MT3DMS format
well_LRCQ_list = create_LRCQ_list(well_obj_list, grid_obj)
ssm_data = create_conc_list(well_obj_list)
#Initialize MODFLOW packages using FloPy
#ml = mf.Modflow(self.name, version='mf2005', exe_name=self.swtexe_name, model_ws=self.dirs[0])
swtm = swt.Seawat(self.name,
exe_name=self.swtexe_name,
model_ws=self.dirs[0])
discret = mf.ModflowDis(swtm,
nrow=grid_obj.nrow,
ncol=grid_obj.ncol,
nlay=grid_obj.nlay,
delr=grid_obj.delr,
delc=grid_obj.delc,
laycbd=0,
top=self.ztop,
botm=self.zbot,
nper=self.nper,
perlen=self.perlen,
nstp=self.nstp,
steady=self.steady)
bas = mf.ModflowBas(swtm,
ibound=grid_obj.IBOUND,
strt=grid_obj.head)
lpf = mf.ModflowLpf(swtm,
hk=self.HK,
vka=self.VK,
ss=0.0,
sy=0.0,
laytyp=0,
layavg=0)
wel = mf.ModflowWel(swtm, stress_period_data=well_LRCQ_list)
words = ['head', 'drawdown', 'budget', 'phead', 'pbudget']
save_head_every = 1
oc = mf.ModflowOc(swtm)
pcg = mf.ModflowPcg(swtm,
mxiter=200,
iter1=200,
npcond=1,
hclose=0.001,
rclose=0.001,
relax=1.0,
nbpol=0)
#ml.write_input()
#Initialize MT3DMS packages
#mt = mt3.Mt3dms(self.name, 'nam_mt3dms', modflowmodel=ml, model_ws=self.dirs[0])
adv = mt3.Mt3dAdv(
swtm,
mixelm=0, #-1 is TVD
percel=1,
nadvfd=1,
#Particle based methods
nplane=0,
mxpart=250000,
itrack=3,
dceps=1e-4,
npl=5,
nph=8,
npmin=1,
npmax=16)
btn = mt3.Mt3dBtn(swtm,
cinact=-100.,
icbund=grid_obj.ICBUND,
prsity=self.PEFF,
sconc=[grid_obj.temp][0],
ifmtcn=-1,
chkmas=False,
nprobs=0,
nprmas=1,
dt0=0.0,
ttsmult=1.5,
ttsmax=20000.,
ncomp=1,
nprs=nprs,
timprs=timprs,
mxstrn=9999)
dsp = mt3.Mt3dDsp(swtm,
al=self.al,
trpt=trpt,
trpv=trpv,
dmcoef=dmcoef)
rct = mt3.Mt3dRct(swtm, isothm=0, ireact=0, igetsc=0, rhob=rho_b)
gcg = mt3.Mt3dGcg(swtm,
mxiter=50,
iter1=50,
isolve=1,
cclose=1e-3,
iprgcg=0)
ssm = mt3.Mt3dSsm(swtm, stress_period_data=ssm_data)
#mt.write_input()
#Initialize SEAWAT packages
# mswtf = swt.Seawat(self.name, 'nam_swt', modflowmodel=ml, mt3dmsmodel=mt,
# model_ws=self.dirs[0])
swtm.write_input()
#Run SEAWAT
#m = mswtf.run_model(silent=True)
m = swtm.run_model(silent=True)
logging.debug('SEAWAT step completed')
#Copy Modflow/MT3DMS output to new files
shutil.copyfile(
os.path.join(self.dirs[0], self.name + '.hds'),
os.path.join(self.dirs[0], self.name + str(period) + '.hds'))
shutil.copyfile(
os.path.join(self.dirs[0], 'MT3D001.UCN'),
os.path.join(self.dirs[0], self.name + str(period) + '.UCN'))
#Create head file object and read head array for next simulation period
h_obj = bf.HeadFile(
os.path.join(self.dirs[0], self.name + str(period) + '.hds'))
grid_obj.head = h_obj.get_data(totim=self.perlen)
#Create concentration file object and read temperature array for next simulation period
t_obj = bf.UcnFile(
os.path.join(self.dirs[0], self.name + str(period) + '.UCN'))
grid_obj.temp = t_obj.get_data(totim=self.perlen)
logging.debug('Output processed')
if well_obj_list:
for i in well_obj_list:
#Update each active Python well object with the temperature and head at its grid location
i.T_modflow = grid_obj.temp[i.L[0], i.R, i.C]
i.H_modflow = grid_obj.head[i.L[0], i.R, i.C]
#Update the NetLogo agents from the corresponding Python objects
write_NetLogo_attriblist(self.netlogo, well_obj_list,
nl_update_well_attribs)
#As an example of data exchange, we can calculate the fraction of the simulated grid in which
#the temperature change is significant, and send this value to a NetLogo global variable
use = subsurface_use(grid_obj, grid_obj.temp)
write_NetLogo_global(self.netlogo, 'SubsurfaceUse', use)
logging.debug('NetLogo update completed')
h_obj.file.close()
t_obj.file.close()
self.netlogo.command('file-close-all')
self._handle_outcomes()
