1 / 29

Enhancing Operational Meteorological Nowcasting with NWC SAF Products

Explore the recent experience of using NWC SAF products for operational nowcasting at AEMET, Spain, focusing on enhancing the display of key meteorological data and improving interpretation methods.

aalvarado
Download Presentation

Enhancing Operational Meteorological Nowcasting with NWC SAF Products

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Recent experience on use of NWC SAF for operational nowcasting purposes at AEMET(Spain) José Miguel Fernández-Serdán ATAP (Forecasting Techniques and Applications Area) AEMET – Agencia Estatal de Meteorología Madrid - Spain

  2. SUMMARY Enhancing/improving use of Meteosat NWC SAF products (and channels) in Operational intranet Page: “Pre-convective environment”* Preliminar use and evaluation of physically-based v2013 NWC SAF subproducts (PCPh and CRPh in PPh/PGE14; Water phase in CT) Other conclusions (*) Based on Presentation at 2014 CWG Workshop www.aemet.es

  3. Enhancing/improving usefulness of Meteosat NWC SAF products (and channels) in Operational intranet Page: “Pre-convective environment” • NWC SAF SPhR products: LI, PW (x3), plus separated guess-model differences. Single display, revised enhancements (2014). • New development as alternative: 3 instability indices, TPW. On complementary display (RGB display using PW products and channel-based). • On further possibilities: • Display of Tendencies: diurnal evolution in LI instability wrt. mean of last few days or climatology. • Use of SPhR profiles (+ new parameters/possibilities) in Page or/and in new system NinJo. www.aemet.es

  4. (Until 2014) operational intranet Page, NWC SAF SPhR products: LI, PW (x3), equivalent display with guess-model differences • Including model fields information in form of switch on/off overlays, and last hour loops. Default colortables. • There is an additional display sub-entry: tendencies. www.aemet.es

  5. Operational in summer 2014: Revised color enhancements, single display with guess-model differences included as overlay • New colorPW: new colortable. LI: only significant or close-to. Guess differences: still unique colortable, “not significant” values not shown, increased contrast. • Graphic equivalent model overlays: possibly no more needed, could be replaced by H.R.model value-added fields (e.g. precipitation or storm probability). www.aemet.es

  6. Alternative: 3 instability indices, TPW • Includes all instability indices provided (normally) in the SPhR: • Slight differences are not only of meteorological significance… • …But they could also help interpreting guess-differences (with some knowledge and additional information) in “satellite terms”. • TPW as most significant (certainly also most accurate), is the only PW information content included in this proposal, layer information to be still shown, … in a different way. KI SHI LI TPW www.aemet.es

  7. Alternative SPhR display 3 instalibity + TPW: Now in Experimental page • To eventually replace PW (3) + LI. • Feedback, characterization, complementary information, needed before operational. www.aemet.es

  8. Ideas as complement (I): RGB display of SPhR PW layer subproducts • Goal: complementary display for ingredients, focus here on convection classification supposing it will initiate. Model wind at least 2 levels (or vertical shear) could be overlaid here instead of on basic display. • Studies just starded. idea by now is SPhR R-G-B = BL-ML-HL (3 “independent” information elements), a 3D (rather rough, vertical) image of water wapor distribution (+ cloudiness). • Work needed: RGB color normalisation, characterization and guidelines for interpretation. BL weight is much related to groud elevation, a possibility is to “elevate” BL (at the expenses of ML) according to ground level up to some 1000m. RGB=BL/ML/HL (each: 2/3 full range) www.aemet.es

  9. Ideas as complement (II): “Airmass” and other RGB channel- based displays RGB (just an early example)= Dif13.4/Dif8.7/10.8 (Diffs.: -5/+5ºC) RGB= Airmass • Another goal, using channels: complementary display for ingredients, now to help understanding relevant guess-difference features in terms of where the information comes from, beyond the rough vertical PW subproducts distribution. • Slightly more than an idea by now. Using window or slightly absorbing channels (as at least IR10.8 and 12.0, 8.7 and 13.4). New system NinJo provides good tools to work on RGB and is suited for investigating and developing this issue. Good guidelines and some expertise will be needed! • Airmass RGB (well known) also includes 3D clear-air information. Already used at synoptic level at 3h intervals, useful at high frequency/resolution? • Likely yes www.aemet.es

  10. Ideas as complement (I): Recent work on RGB display using SPhR PW layer subproducts, adopted criteria • Only BL and ML layers (to increase contrast, reduce colours; HL difficult to use and generally not significant). 2 RGB configurations have been retained (RGB = BL/BL/ML and ML/BL/BL). • Normalization: BL and ML ranges are “expanded” 0-255 according to current BL range (mm) with respect to full range (35/45 mm). • One option of (very simplified) correction for ground elevation has been studied: • The criteria of “uniform vertical mixing ratio” was assumed. • Using Standard atmosphere, (fixed) weight for each slice of 100m in the ML (1500 to 5500 m) is computed. • For a surface elevation file at MSG resolution/projection, (fixed) elevation-dependent weighted sum ancillary mask is computed assuming “uniform base level” at each pixel: • One level (1500-1600 m) is added if 50 m < sfc <= 150 m, • 2 levels (1500 -1700 m) is 150 m < sfc <= 250 m)… • (Up to 11 levels for an elevation >= 1150m). • For each new timeslot and product, weighted sum mask is multiplied to ML PW (mm), and result added/substracted to BL/ML. • Only at that stage is normalization applied. • Other simplifications e.g. uniform RH, could also be studied. These approximations can of course not reproduce the huge variability of WV content in the atmosphere, but the purpose would be here practical, rough and qualitative (time-spatial color contrast analysis). www.aemet.es

  11. Recent work on RGB display using SPhR PW layer subproducts: example RGB = BL/BL/ML (basic, –partly- elev. corrected) www.aemet.es

  12. Recent work on RGB display using SPhR PW layer subproducts: example RGB = ML/BL/BL (basic, –partly- elev. corrected) www.aemet.es

  13. Further possibilities: • Display of subproduct Tendencies: diurnal evolution of LI instability with respect to mean of last few days or climatology. • Use of SPhR profiles (+ new parameters/possibilities) in Page or/and in new system NinJo. • Shows product and, for LI: tendency in 24h and 3h plus –since 6z: variable duration, moment heavily dependent. • Maybe the “most realistic” approach: refer this last to mean at that time of model LI last few days. • NWC SAF SW can be configured to provide profiles. • Forecast profiles (as level-by-level tests) are already input to Op. Fcst Soundings page (see left), and are also being integrated to NinJo (its Sounding Tool has much more possibilities). www.aemet.es

  14. SUMMARY Enhancing/improving use of Meteosat NWC SAF products (and channels) in Operational intranet Page: “Pre-convective environment”* Preliminar use and evaluation of physically-based v2013 NWC SAF subproducts (PCPh and CRPh in PPh/PGE14, Water phase in CT) Other conclusions (*) Based on Presentation at 2014 CWG Workshop www.aemet.es

  15. Stratiform/cumuliform case: “Old” PC/CRR vs. IR/VIS www.aemet.es 15

  16. Stratiform/cumuliform (same): New PCPh/CRPh, comments • Both PCPh/CRPh reduce the noise/presence of stratiform (high level) bands. • PCPh reduces the (too large and certainly unrealistic) extent of low %. • CRPh more/better targetting convection, coherence with PCPh. • “Better” use of (full) range scale (both !?). • Further examples: experimental display of equivalent to operational tool “Multiproduct” (way forward for future alternative display). www.aemet.es 16

  17. (NWC SAF Workshop 2010) NEW CONVECTION SCHEME (II): “MULTIPRODUCT” DISPLAY FOR CHARACTERIZATION OF EXISTING CONVECTION Quick summary of satellite information + reference data (other sources). Integrates: NWCSAF products on convection and precipitation, channel enhancements and combinations, radar and lightning information. All data are switch on/off overlays, wide number of combinations possible. Last + last hour (15min interval). Easy to use but still a few problems e.g. temporal&spatial differences on data, display (ANIS-Java) limitations. Window interface Default: PC+CRR+RDT+lightning Default+EchoTop Default+IR10.8”AEMET” enh. PC>20% 3 classes, dim colours. CRR 4 classes, grey (2)-white- red (>15mm/h). www.aemet.es 17 www.aemet.es

  18. PCPh/CRPh for Multiproduct: display-scale adaptation PC PCPh CRR CRPh (in next examples) CRPh (proposed in experimental page) www.aemet.es 18

  19. Rain at mesoscale, comparison to radar (11:00z) CRPh PCPh PC/CRR Radar EcoTop (RDT and lightning data also shown) www.aemet.es 19

  20. Rain at mesoscale, details NWCSAF/Radar/SEVIRI CMa cloud/snow discrimation limits RGB “Natural” (12:15z) = NIR1.6/HRV/HRV PCPh CRPh Radar coverage “intermittences” (-15/-10min) Cloud related to (low) echotop could be “obscured” by Ci www.aemet.es 20

  21. (…by the way can be shown, same day…) This is definitive display of CMa-derived time-cumulated snow coverage (p.m.) + 2 classes (1st observed/to confirm –blue, Melted –green), overlaying options (daily + ground elevation, a.m., p.m.). Complementary RGB displays (NIR1.6/HRV/HRV + HRV/HRV/IR10.8 minus snow in a roaming window). www.aemet.es 21

  22. PCPh/CRPh too sensitive?, N blocking, 27/01/2015 11:00z Radar PPI (but “Biscay” radar is inactive -here!) IR10.8 CT PCPh CRPh PC (def. enh.) (CRR) www.aemet.es 22

  23. PCPh/CRPh useability cycle (early morning, at Illum. QI >=20%) 11/02/2015 9:00z 9:15z • CMa cloud/snow discrimination seems even slower to react! • Is there already full product coverage of Peninsula at 9:45? • Radar inactive (Madrid -here) could lead to detection errors, but also to false alarms (high-level only echoes from a most distant radar). • And due to complex orography, even under full radar coverage, detection lacks are possible. + Radar PPI, 9:45z www.aemet.es 23

  24. Quite promising, clear improvement. Still to be tested in warm season cases (see also final Other conclusions - Hourly products). Must be also shown or taken into account: Illumination-dependent QI (and snow! - here). Ideas for improvement of experimental (then alternative eventually operational) Multiproduct display: All new/old products available in a first moment. New products as default daytime Ilum. QI >= 20%), old nightime. At 0 < Ilum. QI < 20%, old products with “adapted” display. At the adequate moment, only default products. Final considerations on PCPh/CRPh use www.aemet.es 24

  25. SUMMARY Enhancing/improving use of Meteosat NWC SAF products (and channels) in Operational intranet Page: “Pre-convective environment”* Preliminar use and evaluation of physically-based v2013 NWC SAF subproducts (PCPh and CRPh in PPh/PGE14, Water phase in CT) Other conclusions (*) Based on Presentation at 2014 CWG Workshop www.aemet.es

  26. Value-added of Water Phase as compared to basic CT • The interest is for low, MID (yellow) and high opaque cloudiness. • Extense cloudiness (icing) • Convection at different stages • At central hours of the day, (almost) all well identified (= not broken) cloudiness is classified. www.aemet.es 26

  27. Water Phase dependence of sun illumination 7:00z 11:30z • The product is also available at night/low sun but less useful, most of mid-level cloudiness (even some low level) is classified as “undefined”. • Product shows in any case (at that price) a reasonable night/day continuity. www.aemet.es 27

  28. To be introduced as overlay in Intranet Icing operational (initially experimental) display, for adequate temperature range: First, in addition to operational overlay (works on a simple day or night channel-difference mask). Then as alternative overlay, with now additional class for undefined water phase. At a later stage: study of Use for convection development (“updrafts” related properties), with e.g. RGB displays. Other cloud top properties are of course also of interest, see next. Considerations on Water Fase use www.aemet.es 28

  29. It is a need, to get advantage of product improvement /enhancement / improved validation, for application in operational forecasting at least, to dedicate effort to “User development” (characterization, elaborated user displays). Other products already available to be used are: HRW trajectories and WV winds, RDT Overshooting tops, CRPh 1h accumulations (All data “Hourly precipitation” display). And interest on implementing experimental NWC SAF SW/environment, to evaluate diverse options. 2015-onwards many products/productlines expected of interest e.g.: Cloud properties (reff, LWP), Convection initiation, +1h extrapolations; ASII “evolution” image features. Many thanks! questions, comments? Other conclusions www.aemet.es 29

More Related