ECMWF IFS Rnet (net radiation) Met station Agoufou, 1.5W,15.3N MJJA 2005 weaker seasonal dynamic of Rnet in the model (MJJA mean values rather close) * IFS Rnet larger in May, smaller during the monsoon * also, less day to day variability (just a bit of caution, not exactly same time sampling/smoothing here, + maybe effect of 1 point versus ~(40 km)**2 area) decomposition in SW, LW, up & down fluxes, next slides seasonal dynamics explained mostly by upward fluxes monsoon trends (both SW and LW) more differences between IFS and Met station in 2005 (these plots) than in 2004 (dry year) because of weaker seasonal dynamics of 24-h avg values 3-day running mean Observations provided by the CESBIO team: Mougin, Timouk, Kergoat et al
17
Embed
ECMWF IFS Rnet (net radiation) Met station Agoufou, 1.5W,15.3N MJJA 2005 weaker seasonal dynamic of Rnet in the model (MJJA mean values rather close) *
SW down LW down SW up LW up Met station Agoufou, 1.5W,15.3N MJJA 2005 day of year 2005 ECMWF IFS 24-h avg values 3-day running mean
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ECMWF IFS
Rnet (net radiation)
Met station Agoufou, 1.5W,15.3N
MJJA 2005
weaker seasonal dynamic of Rnet in the model (MJJA mean values rather close)
* IFS Rnet larger in May, smaller during the monsoon* also, less day to day variability (just a bit of caution, not exactly same time sampling/smoothing here, + maybe effect of 1 point versus ~(40 km)**2 area)
decomposition in SW, LW, up & down fluxes, next slidesseasonal dynamics explained mostly by upward fluxes monsoon trends (both SW and LW)
more differences between IFS and Met station in 2005 (these plots) than in 2004 (dry year) because of weaker seasonal dynamics of measured radiative fluxes for dry years
24-h avg values 3-day running
mean
Observations provided by the CESBIO team: Mougin, Timouk, Kergoat et al
Rnet
SWnet
ECMWF IFS
Rnet
SW cloud forcing
clear sky SW net
LWnet
-20-LW cloud forcing
clear sky LW netLWnet
SWnet
Met station Agoufou, 1.5W,15.3N
MJJA 2005 24-h avg values 3-day running
mean
SW dow
n
LW dow
n
SW up
LW up
Met station Agoufou, 1.5W,15.3N
MJJA 2005
day of year 2005 day of year 2005
ECMWF IFS
24-h avg values 3-day running
mean
JunJulAugSep
Met station Agoufou 1.5W,15.3N
2002, Rnet
This slide and next ones for 2003 to 2006 illustrate interannual variability and show how the differences between years take place mostly during daytime
JunJulAugSep
Met station Agoufou 1.5W,15.3N
2003, Rnet
JunJulAugSep
Met station Agoufou 1.5W,15.3N
2004, Rnet
JunJulAugSep
Met station Agoufou 1.5W,15.3N
2005, Rnet
JunJulAugSep
Met station Agoufou 1.5W,15.3N
2006, Rnet
variations along latitude (North Sahel)
weel, with 2 met stations only, 15.3N & 17N, in 2004
interannual variability
with 2004 and 2005, met station at 15.3N
note: there is a meaningful seasonal variation of albedo in the IFS, in terms of shape but it is less than 0.02 in amplitude – I have just drawn a line to indicate the avg value in the IFS.
2004 Bamba, 1.5W , 17.0N
0.4
Albedo ECMWF ~ 0.38
Rnet ECMWF IFS
Rnet Met station
2004 Agoufou 1.5W , 15.3N
0.4
Albedo ECMWF ~ 0.27
Rnet ECMWF IFS
Rnet Met station
2005 Agoufou 1.5W , 15.3N
Albedo ECMWF ~ 0.28
Rnet ECMWF IFS
Rnet Met station
Cloud cover and rainfall in the IFS Sahel, 2 next slides
an example with 1 met station 15.3N , in 2004 24-h average values, except rainfall: 24-h cumulative (also shown, surface heat fluxes H and LE) ok I made them with 24h starting each day something like 6 or 9 hours after the model start, i.e. Possibly a bit early with respect to spin-uo-down but it may be less of a pb at this latitude (?)
then a slide showing some characteristics of the measured rainfall
overall, in the IFS it does not seem to me that there is a clear underestimation of rainfall events but there is an underestimation of rainfall per rainfall event (I base this argument on more than these examples of course)the stratiform rainfall being negligible in the IFS does not seems wrong to me
note: simulated stratiform rainfall negligible
composite precipitation events Agoufou 2002,2003,2005,2006from met station, for Jun Jul Aug Sep, other colours=other months
duration
Rang
e of
fluc
tuat
ions
OBS
ECMWF IFS
LWnet+cstA,SWnet+cstB
Rnet
Distinct interannual signatures of Rnet in OBS & ECMWF IFS, qualitatively similar LW & SW differences for wet (2003) versus dry (2004) monsoon seasons but the sum LW+SW reflect the quantitative differences