atmoschem / vein Goto Github PK

This could be good when for constructing grid with WP in WRF

Needs #30

speciate does not produce mol/h

if
veh <- veh$PC <- c(0, 0, 0, 1, 2, 3 , NA, NA)
0 means that vehicles produced at first and second age (index). Last two NA indicates that the didnt exist vehicles at age 7 and 8. For instance, these type of vehicles started being produced at age 6 but the data already includes the last two NA.

Maybe, it could be a good idea include a filter for excluding NA in this parameter

veh$PC[!is.na(veh$PC)]

According to Writting R Extensions, in DESCRIPTION, the field Depends should be nowadays used rarely, only for packages which are intended to be put on the search path to make their facilities available to the end user, for instance, latticeExtra depends on *lattice. In the case of vein, few times sp functions are called therefore it is not justified.

time in wmis_wrf should use attr(x, "tzone") <- "tz". Therefore, the argument "utc" is not necesary anymore. Also, the name 'wrf' it is no t necessary because it could be used with different models. This function l probably will be deprecated once it is ready the package eix, which will export emissions to different, models, including WRF.

The message needs to separate letters and round the sum of emissions

When running WRF-Chem or other atmospheric models it is necessary meteorological data as input. This data has different temporal resolution and must match with UTC hourly emissions, in general.
A common database is NCAR/UCAR Global Forcast System https://rda.ucar.edu/datasets/ds335.0 but time resolution is each 6 hours. There are other datasets of 3 hours time resolution. Therefore, it would be convenient to have an argument in emis_wrf to match the hours of the emissions to the hours UTC of the meteorological data.

It would be good to include an argument in emis_post that point to a path with a data-frame to rbind newer results. This can wotk like a data-base with newer dependencies, only base.

VEIN reads list of Vehicle data-frame but vkm it is not updated

now message rounds sums of 110^6 vehicles with 2 digits. It should be sum of 110^3 vehicles

los argumentos de emis_post son:

args(emis_post)
function (arra, veh, size, fuel, pollutant, by = "veh") 

veh, size, fuel y pollutant son strings para generar una base de datos clasificada. Sin embargo, podria ser mejor incluir una classificación desde lo mas general a lo mas particular:

• fam (family): PC, LCV, HGV, Bus o MC
• subfam: HGV-A (A de Articulado), HGV-R (R de Rigido), Coach.
• fuel: fuel
• size: cilindradas del motor o peso bruto del camión
• vehcomp: nombre compuesto de la composición vehicular. Por ejemplo: PC_E25_1400.

Generará una base de datos mas pesada pero mejor clasificada.

ed_* emission factors functions are useful but when integrating into a street reality, lacks flexibility. For instance, there are arguments to account for load and gradient but one at the time. It would be good that these functions accept vectorized values, for example:
gradient <- c(0, 0.2, 0.4, 0.6) Representing streets with gradient 0, 0.2, 0.4 and 0.6. The length of this vectorized arguments should be equal to the number of streets. This function should allow having different values at different hours, for example, for evaluating the hourly variation of load in urban buses or trucks.

The age* functions (age_ldv, age_hdv, age_moto and my_age) return a message with a total number of vehicles and average age. It would be good to add an option to shows or not these messages

Missing description of est directory.

Needs #30

emis_wrf should return dataframe with local and UTC time. Local time should be converted based on attr()

emis_posts with by = "veh" assumes a 7 days of the week estimation.

Plots should be spatial and temporal. These plots will be improved after migrating to sf

hours are incorrectly assigned in output of emis_post when veh is a list of vehicles data-frame.

sf is an R package of spatial features. VEIN needs to migrate its dependencies from sp to sf

the updated speciation of IAG excluded CH3OH and KET incorrectly added. This implies to change the function speciate.

normalized hourly profiles are not equal than the resulting profiles after estimating emissions. Hence, it is necessary to add hourly emissions profiles to temporally distribute top-down emissions

This feature would provide a quick check if vehicles distributions are consistent

I was wondering why the emissions profile does not add 24 * 7 = 168.
I always thought that the sum of the emissions of all hours of the day of the week should add up to the total emissions for that week with a uniform profile.
Thanks in advance,
Mel

It would be a good idea print a message with sum of emissions in emis functions

emis_Wrf depends on AAS4WRF (Vera, 2015) configuration. This configuration has changed. Now it is for mechanism CBMZ. The pollutants 36 species. The output of emis_wrf should contain one column per pollutant and in the case that there are no emissions, the column should have filled with 0.
Compounds:

• "E_SO2"
• "E_NO"
• "E_ALD"
• "E_HCHO"
• "E_ORA2"
• "E_NH3"
• "E_HC3"
• "E_HC5"
• "E_HC8"
• "E_ETH"
• "E_CO"
• "E_OL2"
• "E_OLT"
• "E_OLI"
• "E_TOL"
• "E_XYL"
• "E_KET"
• "E_CSL"
• "E_ISO"
• "E_NO2"
• "E_CH3OH"
• "E_C2H5OH"
• "E_PM25I"
• "E_PM25J"
• "E_SO4I"
• "E_SO4J"
• "E_NO3I"
• "E_NO3J"
• "E_ORGI"
• "E_ORGJ"
• "E_ECI"
• "E_ECJ"
• "E_SO4C"
• "E_NO3C"
• "E_ORGC"
• "E_ECC"

It is needed an 'skeleton' function to create folders fir i/o in a known structure, or skeleton. This functons could be named vein::vein and will use dir.create and other file functions.

This functions must be flexible in order to produce folders and/or files.

An EmissionsArray has 4 dimensions, number of streets, the category of vehicles, hours and days. However, as the last two dimensions represent time, maybe it could be better to eliminate the last dimension and the third dimensions could be only the total number of hours.

Sometimes it is necessary to convert to units or unclass several columns of dataframes. In other occasions, it is necessary to add units. It would be convenient to create a function to do that.

VEIN functions could include the dependency on data.table functions so that it produces faster estimations. This could imply in modification in all function but these modifications should be only internal, they must preserve the same arguments

currently Evaporative emissions are based on tier 2 EEA emission guidelines. The improvement includes hourly distribution and Tier 3

The speciation of NMHC from tunnel studie of IAG/USP is incorrect. There is no emissions of methanol. New table will be fixed in next commit

The user must convert the emissions from grams to mol before using emis_wrf. It would be convenient to add a summary of the output when using emis_wrf.

argument name in age functions should have a default value

Emissions are estimated for each type of vehicle, street different hour. Also, emis_post generates
three type of files. At the end there are a lot of files. Hence, an emis_merge function would be useful. This function will not be related with the base::merge functions, but, it will load all emissions according to an specific format and mergte,rbind,cbind them (see #36).

It is important to preserve any spatial attribute, so this function would work better with sf objects (see #5 )

The output of emis_wrf is:
LONG, LAT, ID, ....
and it should be
ID, LONG, LAT, ....

Average Daily Traffic (ADT) is a common metric in transportation studies, used also as input for estimating emissions. It should be included

some mileage functions return negative values some ages.

https://www.researchgate.net/publication/261870923_Curvas_de_intensidade_de_uso_por_tipo_de_veiculo_automotor_da_frota_da_cidade_de_Sao_Paulo?el=1_x_2&enrichId=rgreq-ce74b752-7df5-4db6-a7f2-658fe7d6eb47&enrichSource=Y292ZXJQYWdlOzI2MTg3MDkyMztBUzoxMDEwNDc1MDY4MzM0MTVAMTQwMTEwMzAxMTU2MQ%3D%3D

Light Flex vehicles only

linha 46 e 47 deveria ter deparse para o tipo de veiculo

Age functions allow producing age distribution of vehicles in circulation. The functions are base on scrapping functions of the National Emissions Inventory for Brazil, 1996 and 2001. These functions can be restrictive and need to be updated. The goal is to have functions that can represent on average, a specific vehicular market.

There is an error in table of speciation for painting

Needs #30

I've never used this option

ef_nitro presents an error when filtering for some parameters. The problem is that there are still some values "ALL" that need to be changed at internal data.

This function would allow estimating top-down vehicular emissions. Possible features:

• It allows estimations with mileage by age of use.
• Read spatial polygons with fleet by type of vehicle and/or mileage by age of use.
• If it is used average hourly mileage, it could be used directly inside emis.
• Reads emission factors for n pollutants
• Produces an EmissionsArray which could be processed with emis_post to produce a data-base.

If emis_wrf uses projected emissions grids the output will be projected coordinates. This function automatically should reproject the output to WGS84 lat lon

this can reduce alot of future pŕoblems