runs and dat aset s analysis of t he dat aset s remaining quest - - PowerPoint PPT Presentation

Diurnal cycle of deep convection over land status of the intercomparison case runs and dat aset s analysis of t he dat aset s remaining quest ions & work

use of more realistic surf ace heat f luxes t o improve t he diurnal evolut ion of t he BL 6 SCMs and 3 CRMs runs + addit ional runs (sensit ivit y t est s)

• S. Cheinet also t est ed t his case wit h his

BL mass f lux scheme (quit e encouraging)

runs and datasets

Chaboureau Pet ch CNRM (mésoNH) Met Of f ice (UM) NCAR (UM) model t ype lab (model name) part icipant s Tailleux Pet ch Chaboureau, Koehler, Becht old J ones Piriou ECMWF (I FS) Grenier CNRM (ARPEGE Climat ) SCM LMD (LMDz) SCM SCM Met Of f ice (UM) SMHI (close t o HI RLAM) CNRM (ARPEGE WF) SCM SCM new SCM CRM CRM new CRM Gr abowski

datasets (sever al new it ems) SCMs dat af iles : t ime st ep values (done) Q1 & Q2 : 1 CRMs , 4 CRMs radiat ive f ields : 1 SCM , 2 SCMs convect ive mass f luxes : everybody cloud f ract ion : 3 CRMs & 5 SCMs subgrid scale f luxes and w’ 2 : not a big succes evaporat ion of rain : st ill t o be done a lot of mat erial ! (available on t his comput er)

analysis

again more consist ency among CRMs t han SCMs link bet ween deep convect ion and clouds very weak in some SCMs result s which conf irm what we suspect ed in Ut recht last April, but now wit h t he appropriat e dat aset s some new analysis (moist ure)

night time local solar time (hour)

rainf all events still tend to

• ccurs earlier

in many SCMs than in CRMs with new surf f luxes too

now there is a more realistic diurnal cycle of stability in part icular f or SCMs CI N in SCMs

CRMs

in SCMs

some BLs shallower t han ot hers but most t end t o evolve t oward a moist er BL

Condensation level in SCMs

links wit h BL height dif erences

in CRMs

scat t er increases af t er sunset ,delicat e f or bot h t ypes of models

in SCMs in CRMs

There is a more realist ic diurnal cycle of st abilit y but t he CI N is st ill quasi 0 during t he f ull dayt ime period in SCMs in cont rast wit h CRMs CI N reaches « quasi-0 » around 9h lst in bot h t ypes of models But t he CI N does not explains t he whole st ory in CRMS because rains st art s af t er 12h lst

I n several SCMs, cloud t ops reach high alt it ude direct ly wit hin

• ne t ime st ep

in CRMs

saturation def icit

cloud f raction rain mixing ratio

SCMs: saturation def icit (qv- qvsat) SCMs: upward convective mass f lux

saturation def icit f or the 2 days t his f eat ure is not as dramat ic t he second day but st ill t hrere

dif f erent layerings, probably linked t o dif f erent convect ive t ransport SCMs of en evolve t owards moist er low level condit ions, not enough t ransport ?

same scenario the second day

t o go f urt her: need t o compare Q2 and moist ure f lux (complex)

thetae apparent ly leads t o an at mosphere which is less st able in SCMs t han CRMs

questions raised, propositions made in Utrecht

f actors controlling the length of the non- precipitating phase? moist ure is one f act or impact of downdraf ts on stability: t hey increase t he CI N in CRMs not so clear in SCMs not much impact in some SCMs (because t hey are already t oo weak?) systematic sensitivity study to triggering and downdraf ts? (part ly done, input f rom C. J ones and J .-M. Piriou) GCM tests (part ly done, ECMWF model) SCMs behaviour when convection is turned of f ? how does it compares to the shallow cumulus case? (not been done, H. Grenier moved)

intercomparison paper : in pr eparat ion

act ions: f or now most ly myself wit h some help of J on Pet ch + f or t hose who did not do it yet : inf ormat ion concerning t he run (version, ref erences… ) links bet ween BL, shallow & deep convect ion and cloud propert ies (1/ 2 page t o 1 page) … t alk more at t he end of t he session

wrap- up discussion

timetables, deadlines end of EUROCS: 1 March 2003 submission of papers, special issue: in 2003 (spring or lat t er?) coordinat e t imet ables f or report s & paper s reports, data availability, deliverables report s will include result s f rom CRMs, SCMs and GCMs how long? where do we archive model out put s? available f igures can be put on t he web in J anuary papers, special issue

PAPERS

« Group » paper with a long list of co- authors

includes result s f rom GCMs, RCMS mat erial f or t he int roduct ion, C. J ones & J .-M. Piriou CRMs & SCMs (int ercomparison) discussion rainf all, cloud t op height , st abilit y, t riggering and downdraught s (not in dept h analysis) A quest ion: f or t he int ercomparison, do we dist inguish bet ween models (dif f erent t ypes of curves… ) or not ?

• rganisat ion, your f eedbacks: wit hin 1 mont h, ok?

The « Chaboureau » papers are not f ar f rom ready t he CRM analysis t he CRM-SCM-GCM st udy

The GCI N/ GCAPE paper , Remi Tailleux Papers of Jon Petch & Alison Stirling

• thers?