TOLNet Ozone CMAQ Comparison Plot

This example shows how to compare TOLNet to CMAQ by interpolating CMAQ to the. TOLNet grid. It uses NASA LaRC data during FIREX-AQ as an example with EQUATES CMAQ from pyrsig. You can adapt the example to use your CMAQ on local disk.

Install pytolnet if not avail

# !python -m pip install git+https://github.com/barronh/pytolnet.git

Initialize API and Find Data

import pytolnet
import pyrsig
import pandas as pd
import pyproj
import numpy as np
import matplotlib.pyplot as plt


api = pytolnet.TOLNetAPI()

# Find newest data from UAH
cldf = api.data_calendar('NASA LaRC')
daysdf = cldf.query(
    'start_data_date >= "2019-09-09 00:00:00"'
    + ' and start_data_date < "2019-09-13 00:00:00"'
)

Retrieve TOLNet Data

tdss = []
for id, row in daysdf.iterrows():
    tdss.append(api.to_dataset(id))

Retrieve CMAQ Data

ilat = tdss[0].instrument_latitude   # Use first file for
ilon = tdss[0].instrument_longitude  # instrument location
bbox = (ilon - 0.1, ilat - 0.1, ilon + 0.1, ilat + 0.1)
rsig = pyrsig.RsigApi(bbox=bbox)
qdss = []
for id, row in daysdf.iterrows():
    bdate = pd.to_datetime(row['start_data_date'])
    # Use RSIG to get EQUATES Simulations for comparison
    qds = rsig.to_ioapi('cmaq.equates.conus.conc.O3', bdate=bdate)
    zds = rsig.to_ioapi('cmaq.equates.conus.conc.ZH', bdate=bdate)
    qds['ZH_ASL_MID_KM'] = (zds['ZH'] + qds['ELEVATION']) / 1000.
    # Optional, use local data instead of pyrsig
    # qds = pyrsig.open_ioapi(f'./CCTM_CONC_{bdate:%Y%m%d}')
    # zds = pyrsig.open_ioapi(f'./METCRO3D_{bdate:%Y%m%d}')
    # gds = pyrsig.open_ioapi(f'./GRIDCRO3D_{bdate:%Y%m%d}')
    # qds['ZH_ASL_MID_KM'] = (zds['ZH'] + gds['HT']) / 1000.
    qdss.append(qds)

Interpolate CMAQ to TOLNet

qproj = pyproj.Proj(qdss[0].crs_proj4)  # Use first CMAQ for projection info
ix, iy = qproj(ilon, ilat)
for qds, tds in zip(qdss, tdss):
    # Make a placeholder for interpolated data
    tds['CMAQ_O3'] = tds['derived_ozone'] * np.nan
    tds['CMAQ_O3'].attrs.update(units='ppbv', long_name='CMAQ')
    # Find nearest grid cell and interpolate time
    tsmatch = qds.sel(
        ROW=iy, COL=ix, method='nearest'
    ).interp(TSTEP=tds['time'])
    # For each time-step, interpolate from ZH coordinate to altitude
    for ti, t in enumerate(tsmatch.time):
        ts = tsmatch.sel(time=t)
        # Interpolate along altitude in meters above sea level
        tds['CMAQ_O3'][ti] = np.interp(tds.altitude, ts.ZH_ASL_MID_KM, ts.O3)

Plot a Day

norm = plt.Normalize(30, 70)
cmap = 'viridis'
dlevels = [-35, -25, -15, -5, 5, 15, 25, 35]
dcmap = 'seismic'
gskw = dict(left=0.05, right=0.99)
for tds in tdss:
    sdate = pd.to_datetime(tds.attrs['start_data_date'])
    edate = pd.to_datetime(tds.attrs['end_data_date'])
    loc = tds.attrs['DATA_LOCATION']
    tds = tds.groupby(tds.time.dt.floor('1h')).mean()
    Z1 = tds.derived_ozone.T * 1
    Z1.attrs.update(long_name='TOLNet', units='ppbv')
    Z2 = tds.CMAQ_O3.T.where(~Z1.isnull())
    dZ = Z2 - Z1
    dZ.attrs.update(long_name='CMAQ - TOLNet', units='ppbv')

    fig, axx = plt.subplots(
        1, 3, figsize=(18, 4), sharex=True, sharey=True, gridspec_kw=gskw
    )
    qm = Z1.plot(ax=axx[0], norm=norm, cmap=cmap)
    Z2.plot(ax=axx[1], norm=norm, cmap=cmap)
    dZ.plot(ax=axx[2], levels=dlevels, cmap=dcmap, extend='both')
    axx[0].set(xlim=(sdate, edate))  # show only hours with valid TOLNet data
    plt.setp(axx, facecolor='gainsboro', xlabel='time [UTC]')
    fig.suptitle(loc)
    fig.savefig(f'EQUATES_TOLNet_{loc}_{sdate:%Y-%m-%d}.png')

• 
•