def GenDatelist(start_date,end_date,hour_inc):
try:
st_yr,st_mth,st_day,st_hrs=cal.decompose_date(start_date)
end_yr,end_mth,end_day,end_hrs=cal.decompose_date(end_date)
except ValueError:
print ">>> Need to include hours in date string (YYYYMMDDhh)"
raise ValueError(">>> need more than 3 values to unpack")
start_datetime=dt.datetime(st_yr,st_mth,st_day,st_hrs)
end_datetime=dt.datetime(end_yr,end_mth,end_day,end_hrs)
time_delta=dt.timedelta(0,hour_inc*3600)
cur_datetime=start_datetime
date_list=[]
while cur_datetime < end_datetime:
date_list.append(int(str(cur_datetime.year) + str(cur_datetime.month)\
.zfill(2) + str(cur_datetime.day).zfill(2) + str(cur_datetime\
.hour).zfill(2)))
cur_datetime+=time_delta
return date_list
def select_GDAS_files(date_list,dataset_file_list):
filenames=[]
for date in date_list:
yr,mn,dy,hr=cal.decompose_date(date)
# get the week this day falls in, and the week before
weeks=[dy/7,dy/7+1]
for wk in weeks:
if wk==0:
dataset_filenames=['gdas1.'+cal.month_names[mn-2]+str(yr)[-2:]+'.w5',\
'gdas1.'+cal.month_names[mn-2]+str(yr)[-2:]+'.w4']
else:
dataset_filenames=['gdas1.'+cal.month_names[mn-1]+str(yr)[-2:]+'.w'+str(wk)]
for dataset in dataset_filenames:
if dataset in dataset_file_list:
if dataset not in filenames:
filenames.append(dataset)
else:
raise NameError('required dataset not included in dataset_files:'+dataset)
return filenames
def WriteConcInfile(infile,pollutant_data=None,grid_data=None,deposition_data=None,
date=2006072700,
conc_sourcepts=[-33.1858, 138.0169, 50.0],
run_time=48,
vert_mode=0,
model_top=10000,
# num_datasets=1,
dataset_dir=['/media/sda4/hysplit-data/'],
dataset_name=['gdas1.jul06.w4'],
outfile_dir=['/media/sda4/hysplit-dump/']):
"""function to open and edit hymodelc input file"""
import datetime as dt
# determine total number of sources to be requested
num_sources=len(conc_sourcepts)
# dataset_name may be a string (not iterable) or a list or an array
if hasattr(dataset_name,'__iter__'):
num_datasets=dataset_name.__len__()
else:
num_datasets=1
# same with dataset_dir
if hasattr(dataset_dir,'__iter__'):
num_dataset_dirs=dataset_dir.__len__()
# now, this must be 1 or match the number of datasets
if ((num_dataset_dirs != num_datasets) and (num_dataset_dirs != 1)):
print '>>> Error: Specify one dataset directory OR exactly one directory per dataset!'
return 1
else:
num_dataset_dirs=1
cur_yr,cur_mth,cur_day,cur_hr=cal.decompose_date(date)
cur_dt=dt.datetime(cur_yr,cur_mth,cur_day,cur_hr)
date_str=str(cur_dt.year)[-2:]+' '+str(cur_dt.month).zfill(2)+' '\
+str(cur_dt.day).zfill(2)+' '+str(cur_dt.hour).zfill(2)
# date_str=cal.FormatDate(date)[0]
# open file for write access
fid=open(infile,'w')
print >>fid,date_str
print >>fid,num_sources
for source in conc_sourcepts:
print >>fid, '%3.4f %3.4f %3.1f' % (source[0], source[1], source[2])
print >>fid,run_time
print >>fid,vert_mode
print >>fid,'%3.1f' % (model_top)
print >>fid,num_datasets
for k in range(0,num_datasets):
if (num_dataset_dirs==1):
print >>fid,dataset_dir
else:
print >>fid,dataset_dir[k]
print >>fid,dataset_name[k]
# plume release details, edit here:
if pollutant_data is not None:
num_pollutants=len(pollutant_data['plume_id'])
else:
num_pollutants=1
pollutant_data={}
pollutant_data['plume_id']=['PLU1']
pollutant_data['emission_rate']=[1000000.]
pollutant_data['emission_time']=[96]
pollutant_data['release_start']=[date_str]
print >>fid, num_pollutants
for k in range(num_pollutants):
print >>fid, pollutant_data['plume_id'][k]
print >>fid, pollutant_data['emission_rate'][k]
print >>fid, pollutant_data['emission_time'][k]
print >>fid, pollutant_data['release_start'][k]
# concentration grid
if grid_data is not None:
num_grids=len(grid_data['lat_lon'])
else:
num_grids=1
grid_data={}
grid_data['lat_lon']=[[0.,0.]]
grid_data['spacing']=[[.25,.25]]
grid_data['span']=[[30.,30.]]
grid_data['output_directory']=[['/home/tchubb/pylib/physplit/working/']]
grid_data['output_file']=[['cdump1']]
grid_data['num_vert_levels']=[[5]]
grid_data['vert_levels']=[[100.,200.,500.,1000.,2000.]]
grid_data['sampling_start']=[[date_str + ' 00']]
stop_dt=cur_dt+dt.timedelta(4)
stop_str=str(stop_dt.year)[-2:]+' '+str(stop_dt.month).zfill(2)+' '\
+str(stop_dt.day).zfill(2)+' '+str(stop_dt.hour).zfill(2)
grid_data['sampling_stop']=[[stop_str+ ' 00']]
# grid_data['sampling_stop']=[cal.FormatDate(cal.jump_to_date(date,2))]
grid_data['av_now_max']=[['00','03','00']]
print >>fid, num_grids
for k in range(num_grids):
for member in ['lat_lon','spacing','span','output_directory',\
'output_file','num_vert_levels','vert_levels','sampling_start',\
'sampling_stop','av_now_max']:
next_line=''
for thingy in grid_data[member][k]:
next_line+= str(thingy) + ' '
print >>fid, next_line
# deposition
if deposition_data is not None:
num_depositions=len(deposition_data['resuspension_factor'])
else:
num_depositions=1
deposition_data={}
deposition_data['physical_props']=[[1., 1., 1.]]
deposition_data['dispersion_props']=[[0.006, 0., 0., 0. ]]
deposition_data['deposition_props']=[[1., 3.2E+05, 5.0E-05]]
deposition_data['radioactive_half_life']=[[0.]]
deposition_data['resuspension_factor']=[[0.]]
print >>fid, num_depositions
for k in range(num_depositions):
for member in ['physical_props','dispersion_props','deposition_props',\
'radioactive_half_life','resuspension_factor']:
next_line=''
for thingy in deposition_data[member][k]:
next_line+= str(thingy) + ' '
print >>fid, next_line
fid.close()
return 0
def retrieve_trajectory_archive(date_list, heights=[500,1000,2000], arrival_offset=[-18,-12,-6,0,6,12,18]):
import datetime
from copy import deepcopy
global OUTFILE_ARCHIVE
# initialise dataset structures
traj_archive_keys=[]
for height in heights:
traj_archive_keys.append('traj'+str(height))
traj_arrival_keys=[]
for offset in arrival_offset:
if offset<0:
traj_arrival_keys.append( 'prefront_'+str(abs(offset)).zfill(2)+'h' )
elif offset==0:
traj_arrival_keys.append( 'frontal_'+str(abs(offset)).zfill(2)+'h' )
else:
traj_arrival_keys.append( 'pstfront_'+str(abs(offset)).zfill(2)+'h' )
traj_archive={}
for archive_key in traj_archive_keys:
traj_archive[archive_key]={}
for arrival_key in traj_arrival_keys:
traj_archive[archive_key][arrival_key]=[]
for date in date_list:
yy,mm,dd,hh=cal.decompose_date(date)
event_datetime=datetime.datetime(yy,mm,dd,hh)
retrieval_list=[]
for offset in arrival_offset:
# express arrival offset in seconds
new_datetime=event_datetime+datetime.timedelta(0,offset*3600)
# form date list element
new_date_list_element=''
for date_time_element in [new_datetime.year, new_datetime.month, new_datetime.day,\
new_datetime.hour]:
new_date_list_element+=str(date_time_element).zfill(2)
retrieval_list.append(int(new_date_list_element))
for height in heights:
archive_key='traj'+str(height)
traj_set=harch.GetTrajectories(height=height,time_frame=retrieval_list,\
archive_dir=OUTFILE_ARCHIVE)
traj_set=htools.SortTrajectories(traj_set,'end_date')
add_to_record(traj_set,traj_archive[archive_key])
# traj_set=harch.GetTrajectories(height=500,time_frame=retrieval_list,\
# archive_dir=OUTFILE_ARCHIVE)
# traj_set=htools.SortTrajectories(traj_set,'end_date')
# add_to_record(traj_set,traj500)
#
# traj_set=harch.GetTrajectories(height=1000,time_frame=retrieval_list,\
# archive_dir=OUTFILE_ARCHIVE)
# traj_set=htools.SortTrajectories(traj_set,'end_date')
# add_to_record(traj_set,traj1000)
#
# traj_set=harch.GetTrajectories(height=2000,time_frame=retrieval_list,\
# archive_dir=OUTFILE_ARCHIVE)
# traj_set=htools.SortTrajectories(traj_set,'end_date')
# add_to_record(traj_set,traj2000)
# return traj500, traj1000, traj2000
return traj_archive
