Actris-acsm mailing list

["Prevot Andre (PSI)" <andre.prevot AT psi.ch>]

Dear all

 

 

I would just like to add.

There was always the idea to try to do a consistent analysis of all data sets for the source apportionment as a second paper after the one from Michael Bressi.

 

I would say that we try to focus for now on the same datasets that Michael is using for his overview.

Besides the PMF matrices it would be good to know of the ME-2 results you already obtained so far or already published.

So please send some of this information along or directly to Yuliya Sosedova who will start and hopefully do most of this work if we can find funding for her after spring next year (if you have any spare funding let me know).

 

The main question will be if we can use a consistent analysis at all .. in any case, we would like to try.

 

Michael: can you provide the list of data that you are using for your paper so that people remember which data to send.

 

 

All best from Nanjing

 

Andre

__________________________________________
Paul Scherrer Institut
Dr. Andre Prevot
OFLA/004
5232 Villigen PSI
Schweiz

Telefon: +41 56 310 42 02
E-Mail: andre.prevot AT psi.ch

From: actris-acsm-admin AT lists.psi.ch [mailto:actris-acsm-admin AT lists.psi.ch] On Behalf Of Slowik Jay Gates (PSI)
Sent: Donnerstag, 12. November 2015 12:07
To: actris-acsm
Subject: [Actris-acsm] EETA / PSI PMF of ACSM

 

Hi all,

 

It appears my last email regarding EETA was too cryptic, as some questions have come up regarding what PSI intends to do with the uploaded PMF matrices.  So I’ll try to clarify…

 

This project is the result of discussions among the group at the ACTRIS-ACSM meeting last February (attached to ME-2 workshop).  The main idea here is to use these datasets to improve/validate the AuRo-SoFi (Automatic Rolling SoFi) algorithm to the point where it can be released to the community.  For this we need to take a careful look at a number of long-term (ideally year-long or more) datasets, evaluate various aspects of the algorithm’s stability, compare with conventional SoFi analyses, etc.  In the short term, the idea would be for this to feed into a technical/intercomparison paper focusing on the AuRo-SoFi method/capabilities/inter-comparison.  Of course, we want to be very careful not to compromise the ability of individual groups to separately publish their own results.

 

From our existing work on AuRo-SoFi on PSI datasets, we have the feeling that there are some really interesting things to be learned by this method that could eventually fit into a European overview, perhaps as a follow-up to the overview currently led by Michael Bressi.  But as far as I know this hasn’t been seriously discussed yet and there is anyways a lot of work that would need to be done first – but maybe a possibility for the future…

 

Best,

Jay

__________________________________________
Paul Scherrer Institut
Dr. Jay Gates Slowik
OFLA/006
CH-5232 Villigen PSI

Telefon: +41 56 310 57 55
E-Mail: jay.slowik AT psi.ch