On the assimilation of satellite sounder data in cloudy skies in numerical weather prediction models

et al. 2018

Self Cite

Information about moisture distribution and transportation in the preconvection environment is very important for nowcasting and forecasting severe weather events. The Advanced Himawari Imager (AHI) onboard the Japanese Himawari‐8/‐9 provides high temporal and spatial resolution moisture information useful for weather monitoring and forecasting. Algorithms have been developed for three‐layered precipitable water (LPW: surface to 0.9, 0.9–0.7, and 0.7–0.3 in sigma vertical coordinate) retrievals from AHI infrared band radiances using a Geostationary Operational Environmental Satellite‐R series algorithm working group algorithm. The LPW products from AHI have been validated with in situ measurements. An important application of the AHI LPW product is to improve local severe storm forecasts through assimilating high temporal and spatial resolution moisture information into regional‐ and storm‐scale numerical weather prediction (NWP) models. Assimilation techniques and approaches have been developed; the impact on precipitation forecasts for local severe storm over land from the assimilation of LPWs from AHI shows improvement on heavy precipitation forecasts over those from the assimilation of conventional data. Comparisons between AHI infrared band radiance assimilation and LPW assimilation show overall similar or comparable impact on precipitation forecast. The approaches for assimilating LPW can be applied to the assimilation of data from other advanced imagers such as the Advanced Baseline Imager onboard the U.S. next generation of Geostationary Operational Environmental Satellites‐R series, the Advanced Geosynchronous Radiation Imager onboard the Chinese FengYun‐4 series, and the Flexible Combined Imager onboard the upcoming European Meteosat Third Generation.

Section: Summary and Discussionmentioning

confidence: 99%

Impact of Moisture Information From Advanced Himawari Imager Measurements on Heavy Precipitation Forecasts in a Regional NWP Model

et al. 2018

Self Cite

“…In regions with extensive cloud cover, this approach can drastically reduce the potential impact. Cloud contaminated channels have not been used effectively due to challenges in modeling clouds in both the NWP forecast and radiative transfer models [ Li et al ., ]. This is one reason why NWP analysis and forecasts have larger errors in cloudy regions than in clear skies.…”

Section: Introductionmentioning

confidence: 99%

“…Even with these existing RT models, direct assimilation of IR cloudy radiances is still challenging due to the fact that (1) both NWP and RTM have larger uncertainties in cloudy regions, (2) there is a significant change in the temperature Jacobians at cloud level (assuming level clouds instead of layers), (3) satellite observations and NWP may be inconsistent on clouds (e.g., the satellite has clouds but NWP does not and vice versa), and (4) atmospheric parameters have a higher nonlinearity to the IR radiances in cloudy situations [ Li et al ., ]. Alternative approaches for cloudy radiative transfer calculations are desired in order to conduct intercomparisons between different models, which can lead to better uncertainty quantification and improved applications.…”

Section: Introductionmentioning

confidence: 99%

An efficient radiative transfer model for hyperspectral IR radiance simulation and applications under cloudy‐sky conditions

et al. 2017

Self Cite

An efficient radiative transfer model has been developed for hyperspectral infrared radiance simulation under both clear‐ and cloudy‐sky conditions. The hyperspectral IR cloudy radiative transfer model (HIRTM) combines atmospheric transmittances due to molecular absorption and cloud absorption and scattering from cloud hydrometeors. An efficient analytical Jacobian methodology is also developed under both clear‐ and cloudy‐sky conditions, which is needed both to assimilate cloudy radiances directly into numerical weather prediction models and to retrieve atmospheric soundings and cloud properties simultaneously from cloudy radiance measurements. In comparing HIRTM and its analytical Jacobian with the community radiative transfer model (CRTM), our research has shown that HIRTM's Jacobian calculations are similar to those of CRTM. HIRTM and CRTM synthetic observations derived from model output are compared with corresponding real observations from Geostationary Operational Environmental Satellite 13 Imager observations, and both perform similarly under water clouds, while CRTM is colder than HIRTM for thick ice clouds.

“…Satellite information has become one of the dominant sources for improving numerical weather prediction (NWP) model‐based forecasts. In the last few years, all‐sky satellite‐based microwave sounder data have shown great benefit to forecast improvement at operational NWP systems (Bauer et al, ; Geer et al, ; Zhu et al, ) and in the research community (Li et al, ; Yang et al, ; Zhang et al, ). Infrared (IR) radiances, however, are often not fed into the assimilation system because a large percentage of those data are contaminated with clouds (Eresmaa, ; Wang et al, ).…”

Section: Introductionmentioning

confidence: 99%

“…Typically, only radiances unaffected by clouds are effectively assimilated (McCarty et al, ; McNally, ). Direct assimilation of IR cloudy radiances remains challenging (Li et al, ). Further, the non‐Gaussian distributed errors can degrade the impacts of the data assimilation (Pires et al, ).…”

Section: Introductionmentioning

confidence: 99%

Impacts of Observation Errors on Hurricane Forecasts When Assimilating Hyperspectral Infrared Sounder Radiances in Partially Cloudy Skies

et al. 2019

Self Cite

Hyperspectral infrared (IR) sounders provide high vertical resolution atmospheric sounding information that can improve the forecast accuracy of numerical weather prediction (NWP) models. Due to the challenges of assimilating cloudy radiances, NWP centers usually assimilate only radiances that are not affected by clouds. An imager based cloud‐clearing technique provides an alternative and effective way to remove the cloud effects from a partially cloudy field‐of‐view and derive the equivalent clear sky radiances or the cloud‐cleared radiances (CCRs) for assimilation in NWP. Since the observation error is amplified in the cloud‐clearing, or noise amplification process, it is necessary to inflate the observation errors appropriately in order to achieve the optimal value‐added impact from assimilating CCRs. The estimation of observation error inflation is established and discussed. Hurricane Harvey (2017) and Hurricane Maria (2017) are used to simulate and understand the impacts of observation error inflation on the assimilation of Cross‐track Infrared Sounder CCRs for hurricane forecast improvement. Both the precipitation location and intensity forecasts are improved when assimilating CCRs with an inflated observation error for Hurricane Harvey (2017). Assimilating CCRs with an inflated observation error adjusts the temperature and geopotential height fields and further affects the hurricane structures to improve the hurricane track forecasts, thereby demonstrating the importance of using hyperspectral IR measurements in partially cloudy skies for simulating the hurricane structure and improving its forecast. This method can be applied to other imager/sounder combined observations for improving sounder radiance assimilation in cloudy skies and has potential for operational applications.