Browse

You are looking at 1 - 10 of 18,007 items for :

  • Monthly Weather Review x
  • Refine by Access: Content accessible to me x
Clear All
Jingnan Wang
,
Xiaodong Wang
,
Jiping Guan
,
Lifeng Zhang
,
Tao Chang
, and
Wei Yu

Abstract

The forecast uncertainty, particularly for precipitation, serves as a crucial indicator of the reliability of deterministic forecasts. Traditionally, forecast uncertainty is estimated by ensemble forecasting, which is computationally expensive since the forecast model is run multiple times with perturbations. Recently, deep learning methods have been explored to learn the statistical properties of ensemble prediction systems due to their low computational costs. However, accurately and effectively capturing the uncertainty information in precipitation forecasts remains challenging. In this study, we present a novel spatiotemporal transformer network (ST-TransNet) as an alternative approach to estimate uncertainty with ensemble spread and probabilistic forecasts, by learning from historical ensemble forecasts. ST-TransNet features a hierarchical structure for extracting multiscale features and incorporates a spatiotemporal transformer module with window-based attention to capture correlations in both spatial and temporal dimensions. Additionally, window-based attention can not only extract local precipitation patterns but also reduce computational costs. The proposed ST-TransNet is evaluated on the TIGGE ensemble forecast dataset and Global Precipitation Measurement (GPM) precipitation products. Results show that ST-TransNet outperforms both traditional and deep learning methods across various metrics. Case studies further demonstrate its ability to generate reasonable and accurate spread and probability forecasts from a single deterministic precipitation forecast. It demonstrates the capacity and efficiency of neural networks in estimating precipitation forecast uncertainty.

Open access
Joshua Chun Kwang Lee
,
Javier Amezcua
, and
Ross Noel Bannister

Abstract

Two aspects of ensemble localization for data assimilation are explored using the simplified nonhydrostatic ABC model in a tropical setting. The first aspect (i) is the ability to prescribe different localization length scales for different variables (variable-dependent localization). The second aspect (ii) is the ability to control (i.e., to knock out by localization) multivariate error covariances (selective multivariate localization). These aspects are explored in order to shed light on the cross-covariances that are important in the tropics and to help determine the most appropriate localization configuration for a tropical ensemble–variational (EnVar) data assimilation system. Two localization schemes are implemented within the EnVar framework to achieve (i) and (ii). One is called the isolated variable-dependent localization (IVDL) scheme and the other is called the symmetric variable-dependent localization (SVDL) scheme. Multicycle observation system simulation experiments are conducted using IVDL or SVDL mainly with a 100-member ensemble, although other ensemble sizes are studied (between 10 and 1000 members). The results reveal that selective multivariate localization can reduce the cycle-averaged root-mean-square error (RMSE) in the experiments when cross-covariances associated with hydrostatic balance are retained and when zonal wind/mass error cross-covariances are knocked out. When variable-dependent horizontal and vertical localization are incrementally introduced, the cycle-averaged RMSE is further reduced. Overall, the best performing experiment using both variable-dependent and selective multivariate localization leads to a 3%–4% reduction in cycle-averaged RMSE compared to the traditional EnVar experiment. These results may inform the possible improvements to existing tropical numerical weather prediction systems that use EnVar data assimilation.

Open access
Maziar Bani Shahabadi
and
Mark Buehner

Abstract

Cloud-affected microwave humidity sounding radiances were excluded from assimilation in the hybrid four-dimensional ensemble–variational (4D-EnVar) system of the Global Deterministic Prediction System (GDPS) at Environment and Climate Change Canada (ECCC). This was due to the inability of the current radiative transfer model to consider the scattering effect from frozen hydrometeors at these frequencies. In addition to upgrading the observation operator to RTTOV-SCATT, quality control, bias correction, and 4D-EnVar assimilation components are modified to perform all-sky assimilation of Microwave Humidity Sounder (MHS) channel 2–5 observations over ocean in the GDPS. The input profiles to RTTOV-SCATT are extended to include liquid cloud, ice cloud, and cloud fraction profiles for the simulation and assimilation of MHS observations over water. There is a maximum (35%) increase in the number of channel 2 assimilated MHS observations with smaller increases for channels 3–5 in the all-sky experiment compared to the clear-sky experiment, mostly because of newly assimilated cloud-affected observations. The standard deviation (stddev) of difference between the observed global positioning system radio occultation (GPSRO) refractivity observations and the corresponding simulated values using the background state was reduced in the lower troposphere below 9 km in the all-sky experiment. Verifications of forecasts against the radiosonde observations show statistically significant reductions of 1% in the stddev of error for geopotential height, temperature, and horizontal wind for the all-sky experiment between 72- and 120-h forecast ranges in the troposphere in the Northern Hemisphere domain. Verifications of forecasts against ECMWF analyses also show small improvements in the zonal mean of stddev of error for temperature and horizontal wind for the all-sky experiment between 72- and 120-h forecast ranges. This work was planned for operational implementation in the GDPS in fall 2023.

Open access
Mariko Oue
,
Brian A. Colle
,
Sandra E. Yuter
,
Pavlos Kollias
,
Phillip Yeh
, and
Laura M. Tomkins

Abstract

Limited knowledge exists about ∼100-m-scale precipitation processes within U.S. northeast coastal snowstorms because of a lack of high-resolution observations. We investigate characteristics of microscale updraft regions within the cyclone comma head and their relationships with snowbands, wind shear, frontogenesis, and vertical mass flux using high-spatiotemporal-resolution vertically pointing Ka-band radar measurements, soundings, and reanalysis data for four snowstorms observed at Stony Brook, New York. Updraft regions are defined as contiguous time–height plotted areas with upward Doppler velocity without hydrometeor sedimentation that is equal to or greater than 0.4 m s−1. Most updraft regions in the time–height data occur on a time scale of seconds (<20 s), which is equivalent to spatial scales < 500 m. These small updraft regions within cloud echo occur more than 30% of the time for three of the four cases and 18% for the other case. They are found at all altitudes and can occur with or without frontogenesis and with or without snowbands. The updraft regions with relatively large Doppler spectrum width (>0.4 m s−1) occur more frequently within midlevels of the storms, where there are strong wind shear layers and moist shear instability layers. This suggests that the dominant forcing for the updrafts appears to be turbulence associated with the vertical shear instability. The updraft regions can be responsible for upward mass flux when they are closer together in space and time. The higher values of column mean upward mass flux often occur during snowband periods.

Significance Statement

Small-scale (<500 m) upward motions within four snowstorms along the U.S. northeast coast are analyzed for the first time using high-spatiotemporal-resolution millimeter-wavelength cloud radar pointed vertically. The analysis reveals that updrafts appear in the storms regardless of whether snowbands are present or whether there is larger-scale forcing for ascent. The more turbulent and stronger updrafts frequently occur in midlevels of storms associated with instability from vertical shear and contribute to upward mass flux during snowband periods when they are closer together in space and time.

Open access
Ron McTaggart-Cowan
,
David S. Nolan
,
Rabah Aider
,
Martin Charron
,
Jan-Huey Chen
,
Jean-François Cossette
,
Stéphane Gaudreault
,
Syed Husain
,
Linus Magnusson
,
Abdessamad Qaddouri
,
Leo Separovic
,
Christopher Subich
, and
Jing Yang

Abstract

The operational Canadian Global Deterministic Prediction System suffers from a weak-intensity bias for simulated tropical cyclones. The presence of this bias is confirmed in progressively simplified experiments using a hierarchical system development technique. Within a semi-idealized, simplified-physics framework, an unexpected insensitivity to the representation of relevant physical processes leads to investigation of the model’s semi-Lagrangian dynamical core. The root cause of the weak-intensity bias is identified as excessive numerical dissipation caused by substantial off-centering in the two time-level time integration scheme used to solve the governing equations. Any (semi)implicit semi-Lagrangian model that employs such off-centering to enhance numerical stability will be afflicted by a misalignment of the pressure gradient force in strong vortices. Although the associated drag is maximized in the tropical cyclone eyewall, the impact on storm intensity can be mitigated through an intercomparison-constrained adjustment of the model’s temporal discretization. The revised configuration is more sensitive to changes in physical parameterizations and simulated tropical cyclone intensities are improved at each step of increasing experimental complexity. Although some rebalancing of the operational system may be required to adapt to the increased effective resolution, significant reduction of the weak-intensity bias will improve the quality of Canadian guidance for global tropical cyclone forecasting.

Significance Statement

Global numerical weather prediction systems provide important guidance to forecasters about tropical cyclone development, motion, and intensity. Despite recent improvements in the Canadian operational model’s ability to predict tropical cyclone formation, the system systematically underpredicts the intensity of these storms. In this study, we use a set of increasingly simplified experiments to identify the source of this error, which lies in the numerical time-stepping scheme used to solve the model equations. By decreasing numerical drag on the tropical cyclone circulation, intensity predictions that resemble those of other global modeling systems are achieved. This will improve the quality of Canadian tropical cyclone guidance for forecasters around the world.

Open access
Wei-Ting Fang
,
Pao-Liang Chang
, and
Ming-Jen Yang

Abstract

Intensification of Typhoon Chanthu (2021) along the eastern coast of Taiwan was accompanied by pronounced asymmetry in eyewall convection dominated by wavenumber-1 features, as observed by a dense radar network in Taiwan. The maximum wind speed at 3-km altitude, retrieved from radar observations, exhibited a rapid increase of approximately 18 m s−1 within an 11-h period during the intensification stage, followed by a significant decrease of approximately 19 m s−1 within 8 h during the weakening stage. Namely, Chanthu underwent both rapid intensification (RI) and rapid weakening (RW) within the 24-h analyzed period, posing challenges for intensity forecasts. During the intensifying stages, the region of maximum eyewall convection asymmetry underwent a sudden cyclonic rotation from the eastern to the northern semicircle immediately after the initiation of terrain-induced boundary inflow from the south of the typhoon, as observed by surface station data. This abrupt rotation of eyewall asymmetry exhibited better agreement with radar-derived vertical wind shear (VWS) than that derived from global reanalysis data. This finding suggests that the meso-β-scale VWS is more representative for tropical cyclones than meso-α-scale VWS when the terrain-induced forcing predominates in the environmental conditions. Further examination of the radar-derived VWS indicated that the VWS profile pattern provided a more favorable environment for typhoon intensification. In summary, Chanthu’s RI was influenced by the three factors: 1) terrain-induced boundary inflow from the south of the typhoon, observed by surface station data; 2) low-level flow pointing toward the upshear-left direction; and 3) weak upper-level VWS.

Significance Statement

Tropical cyclone intensity change has been an important issue for both real-time operation and research, but the influence of terrain on intensity change has not been fully understood. Typhoon Chanthu (2021) underwent a significant intensity change near the complex terrain of Taiwan that was observed by a dense radar network. This study analyzes 24 h of radar and weather station data to investigate Chanthu’s evolution. The analyses indicate that the complex terrain affected the low-level flow near the TC. Such a change in flow pattern provided additional boundary inflow and a relatively favorable vertical wind shear pattern for TC intensification.

Open access
Amanda Richter
and
Timothy J. Lang

Abstract

NASA’s Investigation of Microphysics and Precipitation for Atlantic Coast-Threatening Snowstorms (IMPACTS) field campaign gathered data using “satellite-simulating” (albeit with higher-resolution data than satellites currently provide) and in situ aircraft to study snowstorms, with an emphasis on banding. This study used three IMPACTS microwave instruments—two passive and one active—chosen for their sensitivity to precipitation microphysics. The 10–37-GHz passive frequencies were well suited for detecting light precipitation and differentiating rain intensities over water. The 85–183-GHz frequencies were more sensitive to cloud ice, with higher cloud tops manifesting as lower brightness temperatures, but this did not necessarily correspond well to near-surface precipitation. Over land, retrieving precipitation information from radiometer data is more difficult, requiring increased reliance on radar to assess storm structure. A dual-frequency ratio (DFR) derived from the radar’s Ku- and Ka-band frequencies provided greater insight into storm microphysics than reflectivity alone. Areas likely to contain mixed-phase precipitation (often the melting layer/bright band) generally had the highest DFR, and high-altitude regions likely to contain ice usually had the lowest DFR. The DFR of rain columns increased toward the ground, and snowbands appeared as high-DFR anomalies.

Significance Statement

Winter precipitation was studied using three airborne microwave sensors. Two were passive radiometers covering a broad range of frequencies, while the other was a two-frequency radar. The radiometers did a good job of characterizing the horizontal structure of winter storms when they were over water, but struggled to provide detailed information about winter storms when they were over land. The radar was able to provide vertically resolved details of storm structure over land or water, but only provided information at nadir, so horizontal structure was less well described. The combined use of all three instruments compensated for individual deficiencies, and was very effective at characterizing overall winter storm structure.

Open access
Nina Horat
and
Sebastian Lerch

Abstract

Subseasonal weather forecasts are becoming increasingly important for a range of socioeconomic activities. However, the predictive ability of physical weather models is very limited on these time scales. We propose four postprocessing methods based on convolutional neural networks to improve subseasonal forecasts by correcting systematic errors of numerical weather prediction models. Our postprocessing models operate directly on spatial input fields and are therefore able to retain spatial relationships and to generate spatially homogeneous predictions. They produce global probabilistic tercile forecasts for biweekly aggregates of temperature and precipitation for weeks 3–4 and 5–6. In a case study based on a public forecasting challenge organized by the World Meteorological Organization, our postprocessing models outperform the bias-corrected forecasts from the European Centre for Medium-Range Weather Forecasts (ECMWF), and achieve improvements over climatological forecasts for all considered variables and lead times. We compare several model architectures and training modes and demonstrate that all approaches lead to skillful and well-calibrated probabilistic forecasts. The good calibration of the postprocessed forecasts emphasizes that our postprocessing models reliably quantify the forecast uncertainty based on deterministic input information in the form of ECMWF ensemble mean forecast fields only.

Open access
Ryosuke Okugawa
,
Kazuaki Yasunaga
,
Atsushi Hamada
, and
Satoru Yokoi

Abstract

Large amounts of tropical precipitation have been observed as significantly concentrated over the western coast of Sumatra Island. In the present study, we used a cloud-resolving model to perform 14-day numerical simulations and reproduce the distinctive precipitation distributions over western Sumatra Island and adjacent areas. The control experiment, in which the warmer sea surface temperature (SST) near the coast was incorporated and the terminal velocity and effective radius of ice clouds were parameterized to be temperature dependent, adequately reproduced the precipitation concentration as well as the diurnal cycles of precipitation. We then used the column-integrated frozen moist static energy budget equation, which is virtually equivalent to the column-integrated moisture budget equation under the weak temperature gradient assumption, to formulate sensitivity experiments focusing on the effects of coastal SST and upper-level ice clouds. Analysis of the time-averaged fields revealed that the column-integrated moisture and precipitation in the coast were significantly reduced when a cooler coastal SST or larger ice cloud particle size was assumed. Based on the comparison of the sensitivity experiments and in situ observations, we speculate that ice clouds, which are exported from inland convection that is strictly regulated by solar radiation, promote the accumulation of moisture in the coastal region by mitigating radiative cooling. Together with the moisture and heat supplied by the warm ocean surface, they contribute to the large amounts of precipitation here.

Open access
Azusa Takeishi
and
Chien Wang

Abstract

Raindrop formation processes in warm clouds mainly consist of condensation and collision–coalescence of small cloud droplets. Once raindrops form, they can continue growing through collection of cloud droplets and self-collection. In this study, we develop novel emulators to represent raindrop formation as a function of various physical or background environmental conditions by using a sophisticated aerosol–cloud model containing 300 droplet size bins and machine learning methods. The emulators are then implemented in two microphysics schemes in the Weather Research and Forecasting Model and tested in two idealized cases. The simulations of shallow convection with the emulators show a clear enhancement of raindrop formation compared to the original simulations, regardless of the scheme in which they were embedded. On the other hand, the simulations of deep convection show a more complex response to the implementation of the emulators, in terms of the changes in the amount of rainfall, due to the larger number of microphysical processes involved in the cloud system (i.e., ice-phase processes). Our results suggest the potential of emulators to replace the conventional parameterizations, which may allow us to improve the representation of physical processes at an affordable computational expense.

Significance Statement

Formation of raindrops marks a critical stage in cloud evolution. Accurate representations of raindrop formation processes require detailed calculations of cloud droplet growth processes. These calculations are often not affordable in weather and climate models as they are computationally expensive due to their complex dependence on cloud droplet size distributions and dynamical conditions. As a result, simplified parameterizations are more frequently used. In our study we trained machine learning models to learn raindrop formation rates from detailed calculations of cloud droplet evolutions in 1000 parcel-model simulations. The implementation of the developed models or the emulators in a weather forecasting model shows a change in the total rainfall and cloud characteristics, indicating the potential improvement of cloud representations in models if these emulators replace the conventional parameterizations.

Open access