Merge pull request #265 from UW-Hydro/develop

Develop

Author: arbennett

docs/whats-new.rst

@@ -2,6 +2,20 @@
 
 What's New
 ==========
+.. _whats-new.2.4.2:
+
+v2.4.2
+------
+Bug fixes
+~~~~~~~~~
+- Aligning longitude values to be within `[-180, +180]`
+  range to avoid inaccurate time-shifts when `UTC_offset`
+  option is set to `True`
+- Some minor code changes to make MetSim compatible with
+  pandas >v2.0.0
+- When using passthrough option consider shortwave as 
+  full day average rather than daylight average
+
 .. _whats-new.2.4.1:
 
 v2.4.1

metsim/datetime.py

@@ -9,7 +9,7 @@
 
 
 def date_range(start=None, end=None, periods=None, freq='D', tz=None,
-               normalize=False, name=None, closed=None, calendar='standard',
+               normalize=False, name=None, inclusive='both', calendar='standard',
                **kwargs):
     ''' Return a fixed frequency datetime index, with day (calendar) as the
     default frequency
@@ -31,9 +31,8 @@ def date_range(start=None, end=None, periods=None, freq='D', tz=None,
         Normalize start/end dates to midnight before generating date range
     name : str, default None
         Name of the resulting index
-    closed : string or None, default None
-        Make the interval closed with respect to the given frequency to
-        the 'left', 'right', or both sides (None)
+    inclusive : str {“both”, “neither”, “left”, “right”}, default 'both'
+        Include boundaries; Whether to set each bound as closed or open, both as default
     calendar : string
         Describes the calendar used in the time calculations. Default is a the
         standard calendar (with leap years)
@@ -51,7 +50,7 @@ def date_range(start=None, end=None, periods=None, freq='D', tz=None,
     if calendar in ['standard', 'gregorian', 'propoleptic_gregorian']:
         return pd.date_range(start=start, end=end, periods=periods,
                              freq=freq, tz=tz, normalize=normalize, name=name,
-                             closed=closed, **kwargs)
+                             inclusive=inclusive, **kwargs)
     else:
         # start and end are give
         if (start is not None) and (end is not None) and (periods is None):

metsim/disaggregate.py

@@ -18,6 +18,7 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 import itertools
+import math
 from typing import Tuple
 
 import numpy as np
@@ -62,6 +63,10 @@ def disaggregate(df_daily: pd.DataFrame, params: dict,
     df_disagg:
         A dataframe with sub-daily timeseries.
     """
+    # adjust any longitude values to be within [-180, +180] range
+    lon_var = params['domain_vars']['lon']
+    params[lon_var] = math.remainder(params[lon_var], 360)
+
     stop = (df_daily.index[-1] + pd.Timedelta('1 days') -
             pd.Timedelta("{} minutes".format(params['time_step'])))
     dates_disagg = date_range(df_daily.index[0], stop,
@@ -71,6 +76,9 @@ def disaggregate(df_daily: pd.DataFrame, params: dict,
     n_days = len(df_daily)
     n_disagg = len(df_disagg)
     ts = int(params['time_step'])
+    ### assume passthrough implies shortwave was computed using entire day not just daylight like mtclim uses to derive its shortwave
+    if (params['method'] == 'passthrough') & (params.get('sw_averaging', '') != 'daylight'):
+      df_daily['daylength'] = 86400.0
     df_disagg['shortwave'] = shortwave(df_daily['shortwave'].values,
                                        df_daily['daylength'].values,
                                        df_daily.index.dayofyear,
@@ -292,7 +300,7 @@ def prec_TRIANGLE(daily_prec, t_min, prec_peak, prec_dur, ts, do_mix):
         disagg_prec = np.zeros(int(n_days*ts_per_day))
 
         # Loop over days
-        for i, (t, P) in enumerate(daily_prec.iteritems()):
+        for i, (t, P) in enumerate(daily_prec.items()):
 
             if do_mix and t_min[t] < 0:
                 prec_day = P * np.ones(ts_per_day)  / ts_per_day
@@ -637,7 +645,8 @@ def shortwave(sw_rad: np.array, daylength: np.array, day_of_year: np.array,
     if params['method'] == 'mtclim' or params.get('sw_averaging', '') == 'daylight':
         tmp_rad = (sw_rad * daylength) / (cnst.SEC_PER_HOUR * ts_hourly)
     else:
-        tmp_rad = sw_rad * 24
+        ### if passthrough, still want to do this...but rather than using dailylight daylength uses entire day     
+        tmp_rad = (sw_rad * daylength) / (cnst.SEC_PER_HOUR * ts_hourly) #tmp_rad = sw_rad * 24
     n_days = len(tmp_rad)
     ts_per_day = int(cnst.HOURS_PER_DAY * cnst.MIN_PER_HOUR / ts)
     disaggrad = np.zeros(int(n_days * ts_per_day))

metsim/io.py

@@ -287,7 +287,7 @@ def read_netcdf(data_handle, domain=None, is_worker=False,
         if isinstance(ds.indexes['time'], xr.CFTimeIndex):
             ds['time'] = ds.indexes['time'].to_datetimeindex()
         ds['time'] = (ds.indexes['time'] -
-                      pd.Timedelta('11H59M59S')).round('D')
+                      pd.Timedelta(hours=11, minutes=59, seconds=59)).round('D')
 
     if var_dict is not None:
         var_list = list(var_dict.keys())

metsim/metsim.py

@@ -47,6 +47,7 @@
 from dask.diagnostics import ProgressBar
 from netCDF4 import Dataset
 from cftime import date2num
+from datetime import datetime
 
 from xarray.backends.locks import get_write_lock, combine_locks, NETCDFC_LOCK
 
@@ -248,7 +249,7 @@ def _validate_force_times(self, force_times):
                         force_times.values[i]).to_pydatetime()
                 elif '/' in self.params[p]:
                     year, month, day = map(int, self.params[p].split('/'))
-                    self.params[p] = pd.datetime(year, month, day)
+                    self.params[p] = datetime(year, month, day)
                 else:
                     self.params[p] = pd.to_datetime(self.params[p])
 

metsim/tests/test_metsim.py

@@ -12,6 +12,7 @@
 import pandas as pd
 import pytest
 import xarray as xr
+from datetime import datetime
 
 import metsim.cli.ms as cli
 import metsim.metsim
@@ -48,9 +49,9 @@ def __init__(self, config):
                'binary': './metsim/data/state_vic.nc'}
 
 # Dates to run over
-dates = {'netcdf': (pd.datetime(1950, 1, 1), pd.datetime(1950, 1, 31)),
-         'binary': (pd.datetime(1949, 1, 1), pd.datetime(1949, 12, 31)),
-         'ascii': (pd.datetime(1949, 1, 1), pd.datetime(1949, 12, 31))}
+dates = {'netcdf': (datetime(1950, 1, 1), datetime(1950, 1, 31)),
+         'binary': (datetime(1949, 1, 1), datetime(1949, 12, 31)),
+         'ascii': (datetime(1949, 1, 1), datetime(1949, 12, 31))}
 
 # Domain vars
 domain_section = {'netcdf': OrderedDict(lat='lat', lon='lon', mask='mask',