aleip/eealocatorplots

EU-GIRP (EU-Greenhouse gas Inventory Reporting Plots): R program to generate plots from EEA-locator CRF data

eealocatorplots's Issues

Make proper R packages out of the program

See e.g. https://r-pkgs.org/whole-game.html

NIR - Non-key categories

Add a section about non-key categories

NIR agri-chapter - EU-KP

make sure all EU's are EU-KP in the text.

N application ratio check

This needs to be better explained, maybe the calculation made more transparent. The target value it NOT 1, but a value > 1, as otherwise some losses have not been considered. It can be <= 1 if bedding is considered.

Key source categories and flags for issue-files

In step 6 there Key source categories are loaded and flags calculated (to make ticking in the EMRT easier). However, this uses old code and - what is more risky - loads also old files.
The whole part needs to be carefully checked and updated, both in content and programming.

It is disactivated with logicals 'dokeycategories' and 'doaddflags'.

Also we need to see if the outlier files that are generated need all these fields or if it would be more user friendly to reduce the fields (then, they don't have to be calculated).

Improve NIR-agri chapter - policy

• Update text on CAP with new reform
• Include Green Deal and Farm to Fork --> targets!
• Update action plans for NVZs. Use information from the trend issues
• Update on industrial emissions directive?
• Integrated Nutrient Management Plan?
• Other examples retrieved from trend issues.

AD - 3.D.1.3 - Grazing -- AD is missing and IEF average cannot be calculated

Share of liquid and solid manure

hi gema, thanks! i think you are right - this formulae is not correct - needs to be corrected.

From: CARMONA GARCIA Gema (JRC-ISPRA)
Sent: 05 December 2018 16:01
To: LEIP Adrian (JRC-ISPRA)
Subject: CAPRI - Inventory comparison

Hi Adrian,

I am checking a bit how things are calculated in CAPRI for comparison with the inventory data. For emissions from MMS, I think there is a problem with the equations in CAPRI_in_NIR_format.gms (folder 'gams\comparisonplots'). When allocating manure to the MMS, we have:

**Calculation of N excretion per MMS (in kt N/year)

caprinv_allagri2(party,variableUID,InvYears_) $ [UID_TO_ALLTYPE(variableUID,"Liquid") ] = SUM((RALL,MPACT) $ [MAP_IPCC_TO_RALL(party,RALL) and UID_TO_MPACT(variableUID,MPACT) and UID_TO_ALLTYPE(variableUID,"Liquid")] , DATA2(RALL,MPACT,"MANN",InvYears_) * DATA2(RALL,MPACT,"LEVL",InvYears_) * p_nemiDAT_time2(RALL,MPACT,"N","TypeShare","Liquid",InvYears_) * 0.001) ;

caprinv_allagri2(party,variableUID,InvYears_) $ [UID_TO_ALLTYPE(variableUID,"Solid") ] = SUM((RALL,MPACT) $ [MAP_IPCC_TO_RALL(party,RALL) and UID_TO_MPACT(variableUID,MPACT) and UID_TO_ALLTYPE(variableUID,"Solid")] , DATA2(RALL,MPACT,"MANN",InvYears_) * DATA2(RALL,MPACT,"LEVL",InvYears_) * p_nemiDAT_time2(RALL,MPACT,"N","TypeShare","Solid",InvYears_) * 0.001) ;

caprinv_allagri2(party,variableUID,InvYears_) $ [UID_TO_ALLTYPE(variableUID,"GRAZ") ] = SUM((RALL,MPACT) $ [MAP_IPCC_TO_RALL(party,RALL) and UID_TO_MPACT(variableUID,MPACT) and UID_TO_ALLTYPE(variableUID,"GRAZ")] , DATA2(RALL,MPACT,"MANN",InvYears_) * DATA2(RALL,MPACT,"LEVL",InvYears_) * p_nemiDAT_time2(RALL,MPACT,"N","TypeShare","GRAZ",InvYears_) * 0.001) ;

As far as I understand, this considers that Typeshare solid+liquid+grazing= 1. If this is right, it is not correct, because in CAPREG (ammo) we have Typeshare housing + grazing = 1, and housing contains liquid + solid (therefore liquid + solid = 1, they are expressed as a fraction of what is excreted other than on pastures).

Could you please have a look when you have some time and see if I am mistaken?

Thanks,

Gema

Improve NIR-agri chapter: Sentences.

Section 5.2 Emission trends contained the following sentence:

Only emissions from follow the opposite trend, contributing to compensate the emission decrease but with a very low impact (% of agriculture total trend

• Such a sentence doesn't make sense when there is no source category with opposite trend, thus needs to be deleted
• It could then be substituted by a sentence saying that there was no source category with opposite trend
• Requires a check if there is any occurrence to add a sentence or possibly write another one
• This problem might have occurred at several places (undetected).

IEF-range plots

Check if the IPCC default ranges are all correct - and correct if necessary!

file eugirpA.2_meastype.r - a few lines can probably made faster

Check:

• alldata$sector_number<-unlist(lapply(c(1:nrow(alldata)),function(x) gsub("Other \\(please specify\\)","",alldata$sector_number[x])))
• alldata$category<-unlist(lapply(c(1:nrow(alldata)),function(x) gsub("livestock","Livestock",alldata$category[x])))
• alldata$sector_number<-unlist(lapply(c(1:nrow(alldata)),function(x) gsub(alldata$category[x],"",alldata$sector_number[x])))

file: eugirp_aggparentanimal.r

This file calculates the average values for animal types and items that are weighted (by population) .
It is called several times, once for 'swine' and 'sheep', once for 'Dairy Cattle' and 'Non-Dairy Cattle', and once for 'Cattle' (aggregating dairy and non-dairy).
The file loops of meastypes, gases, sources etc. in a quite intransparent way.

• It needs urgently to be improved to make use of data.table features to get more transparent.
• Also, there needs to be a test added if the calculation is OK - we have discovered cases in the past (usually at a point in time where we could not look more carefully)

Aggregation of IEF for Non-Dairy Cattle and Poultry

• Introduce check of final IEF = EM/AD once all IEFs aggregates are calculated.
  In May 2020, for Non-Dairy Cattle and Poultry the EU - IEF obtained from AD and EM deviated slighlty from the one calculated bottom-up from MS IEFs. We haven't checked yet the reason, probably one MS that had an animal type not considered in the aggregation (or otherwise wrongly considered).
  We need a safety check that stops the script when this occurs.

function simplifytestmatrix - called from eugirpA.1_eealocator.r

Lines 27ff the function simpifytestmatrix works only with data.frames. Currently called

• allmethods<-simplifytestmatrix(as.data.frame(allmethods),"year",years2keep)
• allinfos<-simplifytestmatrix(as.data.frame(allinfos),"year",years2keep)
• allnotations<-simplifytestmatrix(as.data.frame(allnotations),"year",years2keep)

Table3.B(a)s2 - missing data for Option C countries

Table3.B(a)s2 data are not complete yet (MCF and CLIMA) - needs to be filled!

Improve NIR-agri chapter: Table 5.3 Key categories for the EU (Agriculture - sector excerpt). Emissions in kt CO2 eq

Table not well formatted.

• Left column (Source categories) needs to be deleted
• Is a double header row + merge of some cells possible in pandoc?
• Otherwise automatic formatting in Word?

Potentially significant issues

This is calculated in Step 4 'Calculating trends and growth rates'.
It uses the agrishares versus Total emissions without LULUCF and applied a maxshare.
However, there are a few questions to be solved, such for what years should it be calculated? Is it suffficient that a category is PSI or does it matter how large the under or overestimation is? I.e if a source category is huge but the potential over/underestimation is very small, then it is not PSI - this should be included in the calculation.
Currently disactivated with 'dontdoforthemoment'.
Requires thinking not only at this point, but also when writing out the individual issues.

function: prepareplot()

This function in file eugirp_funnirplots.r prepares the data for one plot.
It is coded to work with different datasources for comparison plots, and value or ief plots, calculating ranges etc.
The coding is very intransparent with little documentation and likely contains a lot of code that is not required (any more) or is done in a complicated way. It has been adapted to work with data.table for ADEM but not yet for other modes and not for multiple data sources.