BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:4th workshop on assimilating satellite cloud and precipitation obs
 ervations for NWP
DTSTART:20200203T100000Z
DTEND:20200206T160000Z
DTSTAMP:20260415T012200Z
UID:indico-event-146@events.ecmwf.int
CONTACT:events@ecmwf.int
DESCRIPTION:\n\n#ECJCWS2020\n\n \n\nWorkshop description\n\nSatellite obs
 ervations sensitive to cloud and precipitation are key to improving global
 \, regional and convective scale weather forecasts. All-sky assimilation o
 f microwave radiances has been proven through operational assimilation. Fu
 rther\, all-sky infrared and visible radiances are in development\, partic
 ularly to make use of rapidly updated geostationary observations. Cloud an
 d precipitation radar\, cloud lidar and lightning data should all start to
  contribute to NWP in the next decade. However\, the assimilation of cloud
  and precipitation pushes data assimilation beyond its current design assu
 mptions. Models often have state dependent biases in cloudy areas and do n
 ot represent details of the microphysical and sub-grid variability that ar
 e needed to drive the observation operators. These\, based on scattering r
 adiative transfer\, require major simplification to make them fast enough 
 for operational use. The random part of the observation error is dominated
  by the lack of predictability of cloud and precipitation processes. How c
 an we use increasing satellite instrument resolutions when the predictable
  scales of cloud are much lower? Further\, these predictability or represe
 ntation errors are correlated in space and time and likely with the backgr
 ound errors\, but to diagnose these errors is extremely difficult. The pro
 blem is not just correlations\, because cloud and precipitation also give 
 rise to nonlinear\, non-Gaussian bounded processes and errors. To progress
  our ability to use cloud and precipitation data we may need to ‘learn
 ’ better physical models and microphysical assumptions\, improve fast sc
 attering radiative transfer\, and allow data assimilation to handle errors
  that are non-Gaussian\, heterogeneous and heavily correlated.\n\nSessions
  and working groups\n\n1.  Assimilating satellite observations sensitive 
 to cloud and precipitation\n\nWhat are we aiming to get from these observa
 tions? How can we achieve this and what more needs to be done?\n\nThis ses
 sion (and WG) sets the scene\, overviewing the cloud and precipitation inf
 ormation available from satellites as well as the issues in assimilating i
 t. It includes applications from global to storm-scale nowcasting\, and co
 vers the passive microwave\, infrared and visible data sources that are al
 ready starting to be widely exploited\, as well as the emerging use of clo
 ud and precipitation radar and lightning observations. Motivations include
  medium-range forecasting\, initialisation of storms especially for very s
 hort range forecasts\, and improved knowledge of cloud and precipitation p
 rocesses. This session will help introduce some of the cross-cutting issue
 s\, such as the difficulty of dealing with processes that are unpredictabl
 e on small scales\, with large biases between model and observations\, and
  the added difficulty of then measuring the forecast improvements brought 
 by cloud and precipitation sensitive observations. For example\, could all
 -sky assimilation do more damage to the forecasted rain bands of hurricane
 s than it improves them? What kind of verification is needed and in terms 
 of verification data\, are the latest satellite precipitation product inde
 pendent enough of model data to be relevant for being used as reference?\n
 \n2.  Cloud and precipitation modelling\n\nHow can forecast models suppor
 t cloud and precipitation assimilation\, and how can cloud and precipitati
 on assimilation help improve models?\n\nAims to assimilate cloud and preci
 pitation data often fall at the first hurdle if the errors between model a
 nd observations are too great\, or if the model does not represent what th
 e observations actually see. This session and WG looks at existing and fut
 ure microphysics and convection schemes\, and the possible development of 
 multi-moment approaches\, and the unification of microphysical assumptions
  with those in the observation operators. To help improve the forecast mod
 els we may need to start "learning" more from the assimilated observations
 . Is this best done as an external loop (model developers look at O-B stat
 istics and tweak parameters?) or an internal loop\, through parameter esti
 mation or even machine learning?\n\n3.  Observation operators in cloud an
 d precipitation\n\nHow do we go from the forecast model's representation (
 e.g. hydrometeor water content) to what the observations see?\n\nThe obser
 vation operator for cloud and precipitation satellite data is usually RTTO
 V or CRTM\, but many others exist. For all of them\, what are their curren
 t and future capabilities relating to cloud and precipitation? Are they pr
 actical for data assimilation applications\, meaning are they fast enough\
 , are they memory efficient? What capabilities are missing? How do we repr
 esent the often-unknown parameters to which observations are so sensitive\
 , such as size distributions and particle shapes\, overlapping cloud layer
 s\, sub-field-of-view variability and 3D structures? How can we best quant
 ify the errors associated with the fast and approximate methods used for d
 ata assimilation?\n\n4.  Data assimilation methods\n\nHow can data assimi
 lation support greater use of cloud and precipitation observations?\n\nPos
 sible methods for cloud and precipitation assimilation include 4D-Var\, en
 semble Kalman filter\, and particle filters: What are their strengths and 
 limitations when it comes to cloud and precipitation? Which domains are th
 ey suited to? Ideally observations should be assimilated directly as radia
 nces and reflectivities\, but there has been success with 1D-Bayesian appr
 oaches. Also\, 1D-Var and L2 retrieval assimilation continue to be propose
 d. How do we represent background errors in cloud and precipitation and wh
 at do we choose for our control variables? Observation errors are also cri
 tical\, but with errors of representation dominant\, how do we represent i
 nter-channel and inter-observation correlations\, let alone correlations w
 ith the background? Is there a need to represent model errors more explici
 tly\, rather than leaving them part of the observation error? For incremen
 ting cloud and precipitation we also need tangent-linear and adjoint model
 s\, or alternatives such as incrementing operators\, statistical models (s
 uch as implicitly used in ensemble Kalman filters) or even machine learnin
 g approaches.\n\nOrganising committee\n\n•    Alan Geer (ECMWF)\n•
     Niels Bormann (ECMWF)\n•    Thomas Auligné (JCSDA)\n\nScienti
 fic organising committee\n\n•    Stephen English (ECMWF)\n•    R
 ichard Forbes (ECMWF)\n•    Ben Johnson (JCSDA)\n•    Andrew Col
 lard (NCEP)\n•    Christophe Accadia (EUMETSAT)\n•    Philippe C
 hambon (Meteo France)\n•    Christina Köpken-Watts (DWD)\n•    
 Chiara Piccolo (Met Office)\n•    Kozo Okamoto (JMA)\n•    Masah
 iro Kazumori (JMA)\n\n          \n\n \n\nhttps://events.ecmwf.int/
 event/146/
IMAGE;VALUE=URI:https://events.ecmwf.int/event/146/logo-2105863536.png
LOCATION:ECMWF
URL:https://events.ecmwf.int/event/146/
END:VEVENT
END:VCALENDAR