Pre-processing for nei2017gb17j emissions

@ctena , we need to create a pre-proc script to adapt the original nei2017gb17j emissions to the HERMESv3 input requirements. The idea is to produce daily emissions.

Original files are available here:

/esarchive/recon/epa/nei2017gb17j/original_files/ag

For now only the agricultural sector is available, but in the future we may add additional sectors.

Original emissions are reported at hourly resolution, 1 file per day with 25 time steps in each file, in moles/s. Emissions should be converted to kg/s-1.m-2. For that, the following steps need to be followed:

1. for each day (each file) compute average hourly emissions considering the first 24 time steps (you can ignore the last time step).

2. For each species, convert from moles/s to kg/s multiplying them for their corresponding molecular weight (here in the table reported in g/mol, so additionally you will need to divide by 1000):

Model Species	Definition	Molecular Weight	Explicit or Lumped
AACD	acetic acid	60	E
ACET	acetone	58.1	E
ALD2	acetaldehyde	44	E
ALD2_PRIMARY	acetaldehyde from emissions only	44	E
ALDX	aldehydes with 3 or more carbons	58.1	L
APIN	alpha pinene	136.2	E
BENZ	benzene	78.1	E
NH3	ammonia	17.0	E
NH3_FERT	ammonia fertilizers	17.0	E
MEOH	methanol	32	E
ETH	ethene	28	E
ETHA	ethane	30.1	E
ETHY	ethyne (acetylene)	26	E
ETOH	ethanol	46.1	E
FACD	formic acid	46	E
FORM	formaldehyde	30	E
IOLE	internal alkene bond	56.1	L
ISOP	isoprene	68.1	E
KET	carbon-ketone bond	72.1	L
OLE	terminal alkene bond	42.1	L
PAR	carbon-carbon single bond	72.1	L
SOAALK	tracer for alkanes that can form secondary organic aerosol	112	L
TERP	monoterpenes	136.2	L
TOL	toluene and other monoalkyl aromatics	92.1	L
FORM_PRIMARY	formaldehyde from emissions only	30	E
NAPH	naphthalene	128.2	E
XYLMN	xylene and other polyalkyl aromatics except naphthalene	106.2	L

3. For each specie, divide emission by grid cell area (in m2). For that, you will first need to define the grid of the domain. You can create it using the temporal_coords produced by HERMES considering the following description of lambert conformal domain:

if domain_type == lcc:
    lat_1 = 33.0 
    lat_2 = 45.0
    lon_0 = -97.0
    lat_0 = 40.0
    nx = 459
    ny = 299
    inc_x = 12000
    inc_y = 12000
    x_0 = -2556000.0
    y_0 = -1728000.0

The final name of the species can be the same as now but in lowercase (e.g. nh3_fert instead of NH3_FERT)

In the script you create, you can mention that the original emissions are downloaded from the following link: https://dataverse.unc.edu/dataset.xhtml?persistentId=doi:10.15139/S3/TCR6BB

Let me know if you have any doubt. thanks!