Search

The standard approach for assimilating satellite radiance observations is to interpolate all vertical levels of the background state and analysis increment to the same horizontal location for input to the radiative transfer model. This can add significant error for observations with large zenith angle. The impact of accounting for the true slanted satellite-viewing geometry was tested by modifying the horizontal interpolation routines in Environment and Climate Change Canada’s Global Deterministic Weather Prediction System. Consequently, model variables are interpolated to a different horizontal position at each model level, for either just the innovation or both the innovation and increment calculation. When this slant-path operator is used for simulation of radiances, reductions in innovation standard deviation, up to 4.5%, for upper-tropospheric and stratospheric temperature and humidity channels of ATMS, AMSU-A, MHS, and CrIS instruments have similar magnitudes as reported in previous studies. In data assimilation experiments, statistically significant reductions in innovation standard deviation (up to 0.3%) for global GPSRO observations are obtained, due to an improved background state. Verification of short- and medium-range forecasts against ERA5, and our own analyses over the region poleward of 60°S show statistically significant reductions of error standard deviation by 2%–3% for wind and temperature in the upper troposphere and lower stratosphere. These positive impacts are mostly due to performing slant-path interpolation on the background state, while also using slant path on the analysis increment has little additional impact. This is expected since the analysis increment in this global configuration has lower spatial resolution, with grid spacing comparable with the maximum horizontal position error from not using the slant-path operator.

A new ensemble-based land surface data assimilation (DA) system is coupled with the atmospheric four-dimensional ensemble-variational data assimilation (4D-EnVar) system with the goal of improving the analyses within Environment and Climate Change Canada’s Global Deterministic Prediction System. Since 2001, the sequential assimilation of surface variables is used to generate the initial conditions to launch the Global Environmental Multiscale (GEM) coupled forecast model. The work presented here is to replace the sequential DA with an independent surface DA system, the Canadian Land Data Assimilation System (CaLDAS) assimilating screen-level observations, and to compare assimilation experiments with CaLDAS run in uncoupled and weakly coupled modes. In the uncoupled mode, CaLDAS is used to initialize the forecast without interacting with the 4D-EnVar system. In the coupled mode, the analyses generated from CaLDAS and 4D-EnVar are used to initialize the forecast model. The analyses and forecasts from uncoupled and coupled runs are evaluated against surface and radiosonde observations over different subdomains to conclude the impact of coupling CaLDAS with 4D-EnVar. Results indicate a statistically significant reduction in bias and standard deviation at the surface for screen-level temperature and dewpoint temperature on the order of 0.1 K, and in the lower troposphere between 1000 and 500 hPa on the order of 0.1 dam for geopotential height and 0.1 K for air temperature and dewpoint depression in the coupled DA runs. The positive impact persists up to 5 days over some subdomains. It is concluded that the coupled DA approach generally performs better than the uncoupled version.

Continuous observation of polar regions from space remains an important unsolved technical challenge of great interest for the international meteorological community. This capacity would allow achieving global continuous coverage once combined with the geostationary (GEO) satellite network. From a practical point of view, continuous coverage of polar regions with a small number of spacecraft can be obtained from a constellation of satellites either in highly elliptical orbits (HEO) or in medium Earth orbits (MEO). The study compares HEO and MEO satellite constellations for their capacity to provide continuous imaging of polar regions as function of the viewing zenith angle (VZA) and evaluates the corresponding latitude limits that ensure sufficient overlap with GEO imagery. Earlier studies assumed the latitude boundary of 60° and the VZA range 70°–85° depending on the space mission focus: meteorological purposes or communications. From the detailed analysis of meteorological retrieval requirements, this study suggests that the overlap of the GEO and polar observing systems (HEO or MEO) should occur down to the latitude band 45°–50° with a maximum VZA ranging between 60° and 64°. This coverage requirement can be met with two sets of three-satellite HEO constellations (one for each polar area) or a six-satellite MEO constellation. The 12-h Molniya and 14-, 15-, and 16-h HEO systems have been analyzed and determined to meet these revised requirements. The study demonstrates that the six-satellite 24-h MEO system can provide a suitable solution, which is also beneficial from the point of view of ionizing radiation and image acquisition geometry. Among the HEO systems, the 16-h HEO has some advantages relative to other HEO systems from the point of view of spatial coverage and space radiation.

Slant satellite-viewing geometry is investigated for the simulation and assimilation of radiances in Environment Canada’s weather forecast system. The standard approach is to extract from a short-term forecast (trial field) a one-dimensional vertical profile, located at the ground footprint of the observation location, to compute the model equivalent of the observation. Since in general, the lines of sight are not vertical, the observation operator can be improved by interpolating the trial field to the slant path. The interpolation relies on the estimation of horizontal gradients of atmospheric variables, extracted at each altitude layer, and representative of a neighborhood of approximately 100 km. Using these slant profiles of the trial fields yields significant improvements in the simulation and assimilation of radiance observations. The impact of this modification in assimilation and forecasts is evaluated over two periods of two months. Both short-term and long-term effects are found, obtained not only from the more accurate interpolation of the trial fields, but also from a revision of the observation error suggested by the lower observation minus trial field difference statistics. The error standard deviation of the geopotential height shows, in the upper troposphere and lower stratosphere, in 12–48-h forecasts, up to 10% reduction in the zonal value at higher latitudes, and 1%–2% reduction in the global value. Given these improved statistics, a revision of the radiance error statistics was explored. Applying these revised statistics suggests that the impact of a slant operator extends into longer range.

A posteriori consistency diagnostics have been used in recent years to estimate correlated observation error. These diagnostics provide an estimate of what the observation error covariances should be and could, in turn, be introduced in the assimilation to improve the statistical consistency between the error statistics used in the assimilation and those obtained from observation departures with respect to the background and the analysis. To estimate the observation error covariances, it is often assumed that the background error statistics are optimal, an assumption that is open to criticism. The consequence is that if the background error covariances are in error, then the estimated observation error statistics will adjust accordingly to fit the innovation error covariances. In this paper, the RTTOV radiative transfer model is used as the observation operator. Using controlled experiments, the background error is considered fixed, and it is shown that the iterative procedure to estimate the observation error may require more than one iteration. It is also shown that the underlying matrix equation being solved can be factorized, and the exact solution can be obtained. If the true background error covariances are used in the assimilation, the estimated observation error covariances are then obtained by subtracting the background error covariances from those of the innovations. This can be applied to the full set of assimilated observations. Using the Environment Canada assimilation system, the results for several types of observations indicate that the background error estimation would deserve additional attention.

This study highlights recent progress at the Canadian Meteorological Centre in the assimilation of Atmospheric Infrared Sounder (AIRS) and Infrared Atmospheric Sounding Interferometer (IASI) radiance observations that are sensitive to land surfaces. The assimilation is carried out using the Canadian global ensemble–variational system. That system benefits from flow-dependent background-error statistics that include covariances between surface skin temperature and atmospheric variables. Up to 142 channels from both AIRS and IASI are assimilated. A detailed database of spectral surface emissivity is used. Forecasts are evaluated against ERA-Interim analyses, own-cycle analyses, and radiosonde observations. A conservative approach is taken, with restrictive conditions on topography and surface emissivity. From 2-month assimilation experiments, a significant positive impact is obtained in the Northern Hemisphere extratropics beyond day 2 from the surface to the upper troposphere. The impact is mixed in other regions, depending on level or forecast lead time. Causes for these mixed results are examined in view of future experiments and eventual operational implementation.

A novel type of multiple-apogee highly elliptical orbits termed as MAP HEO with a period of rotation between 14 h and 15 h is introduced. These orbits are designed to achieve continuous geostationary (GEO)-like imaging of the polar regions in an optimum way. The combination of GEO and HEO satellites would then offer continuous monitoring of weather from space at any point of the globe. This capacity would represent a breakthrough for short- and long-term weather forecasting and narrowing uncertainties in the knowledge of the Earth’s climate through better sampling and more accurate characterization of the diurnal cycle. MAP HEO systems can be launched at critical inclination and are characterized by a local minimum of ionizing radiation. These features simplify the process of orbit maintenance, reduce radiation shielding requirements, and favor a longer lifetime of the mission. Unlike previously considered HEO systems implemented for communications, such as 12-h Molniya and 24-h Sirius radio systems, a MAP HEO constellation achieves a uniform geometrical sampling, which reduces view angle dependent biases. These observational conditions with complete coverage of the diurnal cycle, diverse range of solar illumination, and viewing observational conditions are beneficial for high-latitude meteorological and climate applications, such as the retrieval of Essential Climate Variables (ECV).

A major set of changes was made to the Environment Canada global deterministic prediction system during the fall of 2014, including the replacement of four-dimensional variational data assimilation (4DVar) by four-dimensional ensemble–variational data assimilation (4DEnVar). The new system provides improved forecast accuracy relative to the previous system, based on results from two sets of two-month data assimilation and forecast experiments. The improvements are largest at shorter lead times, but significant improvements are maintained in the 120-h forecasts for most regions and vertical levels. The improvements result from the combined impact of numerous changes, in addition to the use of 4DEnVar. These include an improved treatment of radiosonde and aircraft observations, an improved radiance bias correction procedure, the assimilation of ground-based GPS data, a doubling of the number of assimilated channels from hyperspectral infrared sounders, and an improved approach for initializing model forecasts. Because of the replacement of 4DVar with 4DEnVar, the new system is also more computationally efficient and easier to parallelize, facilitating a doubling of the analysis increment horizontal resolution. Replacement of a full-field digital filter with the 4D incremental analysis update approach, and the recycling of several key variables that are not directly analyzed significantly reduced the model spinup during both the data assimilation cycle and in medium-range forecasts.

There is a well-recognized spatiotemporal meteorological observation gap at latitudes higher than 55°, especially in the region 55°–70°. A possible solution to address this issue is a constellation of four satellites in a highly elliptical orbit (HEO), that is, two satellites for each polar region. An important satellite product to support weather prediction is atmospheric motion wind vectors (AMVs). This study uses observing system simulation experiments (OSSEs) to evaluate the benefit to forecasts resulting from the assimilation of HEO AMVs covering one or both polar regions. The OSSE employs the operational global data assimilation system of the Canadian Meteorological Center. HEO AMVs are assimilated north of 50°N and south of 50°S. From 2-month assimilation cycles, the study examines the following three issues: 1) the impact of AMV assimilation in the real system, and how this compares to the impact seen in the simulated system, 2) the added value of HEO AMVs in the Arctic on top of what is currently available, and 3) the relative impact of HEO AMVs in the Arctic and Antarctic in comparison with no AMVs. Although the simulated impact of currently available AMVs is somewhat higher than the real impact, a firm conclusion is that the added value of Arctic HEO AMVs is substantial, improving predictability at days 3–5 by a few hours in terms of 500-hPa geopotential height. The impact of HEO AMVs is relatively stronger in the Southern Hemisphere. Forecast validation of atmospheric profiles against the simulated “true” state and against analyses generated within the assimilation cycles yields very similar results beyond 48 h.

The Observing System Simulation Experiment (OSSE) capability developed at Environment Canada allows simulation of all observation types currently used operationally as well as future data types. Its infrastructure, based on the operational global data assimilation system used at the Canadian Meteorological Centre, was recently enhanced to conduct data assimilation experiments for two future satellite missions. This study presents a subcomponent of that system, focusing on the assimilation of infrared radiances from the Atmospheric Infrared Sounder (AIRS) and Infrared Atmospheric Sounding Interferometer (IASI) instruments. The goal is to realistically simulate the radiance observations and to reproduce statistical characteristics of these data seen in the real system, notably background and analysis departures. Care is taken to emulate the operational quality control procedures leading to the assimilation of clear radiances, which implies radiance simulation for all-sky conditions. It is found that the standard deviation of the Gaussian random perturbation applied to the simulated observations should be close to that of the radiometric noise level for sounding channels in the 13.0–14.5-μm region but that it should be significantly higher for water vapor channels in the 5.5–6.7-μm region. The study also allows evaluation of residual biases linked to cloud contamination. For atmospheric window channels, that bias can reach −0.4 K. It reduces rapidly with peak height of the channel response function. This suggests that improvements are needed in low-cloud detection. The realism of the OSSE is further demonstrated through the shown impact consistency of AIRS and IASI radiance assimilation in forecasts up to 5 days from the separate use of simulated and real observations.

A new system that resolves the stratosphere was implemented for operational medium-range weather forecasts at the Canadian Meteorological Centre. The model lid was raised from 10 to 0.1 hPa, parameterization schemes for nonorographic gravity wave tendencies and methane oxidation were introduced, and a new radiation scheme was implemented. Because of the higher lid height of 0.1 hPa, new measurements between 10 and 0.1 hPa were also added. This new high-top system resulted not only in dramatically improved forecasts of the stratosphere, but also in large improvements in medium-range tropospheric forecast skill. Pairs of assimilation experiments reveal that most of the stratospheric and tropospheric forecast improvement is obtained without the extra observations in the upper stratosphere. However, these observations further improve forecasts in the winter hemisphere but not in the summer hemisphere. Pairs of forecast experiments were run in which initial conditions were the same for each experiment but the forecast model differed. The large improvements in stratospheric forecast skill are found to be due to the higher lid height of the new model. The new radiation scheme helps to improve tropospheric forecasts. However, the degree of improvement seen in tropospheric forecast skill could not be entirely explained with these purely forecast experiments. It is hypothesized that the cycling of a better model and assimilation provide improved initial conditions, which result in improved forecasts.

A highly elliptical orbit (HEO) with a 16-h period is proposed for continuous meteorological imaging of polar regions from a two-satellite constellation. This orbit is characterized by three apogees (TAP) separated by 120°. The two satellites are 8 h apart, with repeatable ground track in the course of 2 days. Advantages are highlighted in comparison to the Molniya 12-h orbit described in detail in a previous study (Trishchenko and Garand). Orbital parameters (period, eccentricity, and inclination) are obtained as a result of an optimization process. The principles of orbit optimization are based on the following four key requirements: spatial resolution (apogee height), the altitude of crossing the trapped proton region at the equator (minimization of radiation doze caused by trapped protons), imaging time over the polar regions, and the stability of the orbit, which is mostly defined by the rotation of perigee. The interplay between these requirements points to a 16-h period with an eccentricity of 0.55 as the optimum solution. The practical range of orbit inclinations that could be maintained during the spacecraft lifetime can vary from a critical value of 63.435° to 70° (subject to the amount of propellant available for orbital maneuvers). In comparison to Molniya, this type of orbit reduces the radiation exposure to high-energy protons by factor of 103–104. On the other hand, the main advantage of 16 h versus longer orbital periods up to 24 h is better spatial resolution as a result of a lower apogee height. A two-satellite TAP constellation with an orbital inclination of 66° provides 100% temporal coverage above 60°N, >95% above 55°N, >85% above 50°N, and >75% above 45°N.

There has been a significant increase of interest in the building of a comprehensive Arctic observing system in recent years to properly and timely track the environmental and climate processes in this vast region. In this regard, a satellite observing system on highly elliptical orbit (HEO) with 12-h period (Molniya type) is of particular interest, because it enables continuous coverage of the entire Arctic region (58°–90°N) from a constellation of two satellites. Canada is currently proposing to operate such a constellation by 2017. Extending the pioneering study of S. Q. Kidder and T. H. Vonder Haar, this paper presents in-depth analysis of spatiotemporal sampling properties of the imagery from this system. This paper also discusses challenges and advantages of this orbit for various applications that require high temporal resolution and angular sampling.

The interchannel observation error correlation (IOEC) associated with radiance observations is currently assumed to be zero in meteorological data assimilation systems. This assumption may lead to suboptimal analyses. Here, the IOEC is inferred for the Atmospheric Infrared Radiance Sounder (AIRS) hyperspectral radiance observations using a subset of 123 channels covering the spectral range of 4.1–15.3 μm. Observed minus calculated radiances are computed for a 1-week period using a 6-h forecast as atmospheric background state. A well-established technique is used to separate the observation and background error components for each individual channel and each channel pair. The large number of collocations combined with the 40-km horizontal spacing between AIRS fields of view allows robust results to be obtained. The resulting background errors are in good agreement with those inferred from the background error matrix used operationally in data assimilation at the Meteorological Service of Canada. The IOEC is in general high among the water vapor–sensing channels in the 6.2–7.2-μm region and among surface-sensitive channels. In contrast, it is negligible for channels within the main carbon dioxide absorption band (13.2–15.4 μm). The impact of incorporating the IOEC is evaluated from 1D variational retrievals at 381 clear-sky oceanic locations. Temperature increments differ on average by 0.25 K, and ln(q) increments by 0.10, where q is specific humidity. Without IOEC, the weight given to the observations appears to be too high; the assimilation attempts to fit the observations nearly perfectly. The IOEC better constrains the variational assimilation process, and the rate of convergence is systematically faster by a factor of 2.

This paper makes use of ensemble forecasts to infer the correlation between surface skin temperature Ts and air temperature Ta model errors. The impact of this correlation in data assimilation is then investigated. In the process of assimilating radiances that are sensitive to the surface skin temperature, the Ts–Ta error correlation becomes important because it allows statistically optimal corrections to the background temperature profile in the boundary layer. In converse, through this correlation, surface air temperature data can substantially influence the analysis of skin temperature. One difficulty is that the Ts–Ta correlation depends on the local static stability conditions that link the two variables. Therefore, a correlation estimate based on spatial or temporal averages is not appropriate. Ensembles of forecasts valid at the analysis time provide a novel means to infer the correlation dynamically at each model grid point. Geostationary Operational Environmental Satellite (GOES)-8 and -10 surface-sensitive imager radiances are assimilated with and without the inferred correlations in a 3D variational analysis system. The impact of the correlation on analyses is assessed using independent radiosonde data. The impact on 6-h forecasts is also evaluated using surface synoptic reports. The influence of the correlation extends from the surface to about 1.5 km. Temperature differences in the resulting analyses on the order of 0.3–0.6 K are typical in the boundary layer and may extend over broad regions. These difference patterns persist beyond 6 h into the forecasts.

Geostationary Operational Environmental Satellite (GOES)-East and -West window channel radiances are directly assimilated using a 1D variational technique, providing surface skin temperature (Ts) estimates over all surface types (land, water, or ice) from a unique system. This is an important advantage over commonly used regression methods, such as split window. The physical nature of the method allows any combination of channels to be used, and adaptation to new sensors is straightforward. A full month (May 2001) of GOES-8 and -10 data is processed every 6 h; Ts estimates are obtained using radiances from imager channels 4 (11 μm) and 5 (12 μm). Imager channel 2 (3.9 μm) can also be used at night. Surface emissivity maps were constructed from available information based on surface type. The diurnal cycle is studied; its range is on the order of 0.7 K over the ocean. Over land, the diurnal range reaches 30 K for mountainous regions, such as the Rockies or Andes. A full GOES disk image can be proc...

To improve cloud and precipitation forecasts in the Canadian numerical weather prediction system, three cloud schemes with various degrees of complexity are evaluated. Several winter cases are simulated and mesoscale forecasts are compared with satellite, radiosonde, and surface observations. In particular, the distribution of cloud-top pressure is studied in detail using Geostationary Operational Environmental Satellite-8 (GOES-8) data. The three-dimensional structure of temperature and moisture is compared with upper-air data and forecasts of surface precipitation are scored against synoptic observations. This study demonstrates that the operational Canadian cloud scheme has problems that can be partially addressed by the inclusion of more realistic cloud microphysics in the forecast system.

An intercomparison of radiation codes used in retrieving upper tropospheric humidity (UTH) from observations in the v2 (6.3 microns) water vapor absorption band was performed. This intercomparison is one part of a coordinated effort within the GEWEX Water Vapor Project (GVaP) to assess our ability to monitor the distribution and variations of upper tropospheric moisture from space-borne sensors. A total of 23 different codes, ranging from detailed line-by-line (LBL) models, to coarser resolution narrow-band (NB) models, to highly-parameterized single-band (SB) models participated in the study. Forward calculations were performed using a carefully selected set of temperature and moisture profiles chosen to be representative of a wide range of atmospheric conditions. The LBL model calculations exhibited the greatest consistency with each other, typically agreeing to within 0.5 K in terms of the equivalent blackbody brightness temperature (T(sub b)). The majority of NB and SB models agreed to within +/- 1 K of the LBL models, although a few older models exhibited systematic T(sub b) biases in excess of 2 K. A discussion of the discrepancies between various models, their association with differences in model physics (e.g. continuum absorption), and their implications for UTH retrieval and radiance assimilation is presented.

A continuous assimilation of high-density global satellite observations is required in order to improve numerical weather prediction analyses used to start forecasts. Until now, it was assumed that efficiency requirements imposed the use of regression-based models of atmospheric transmittance (typically on fixed pressure layers with coefficients varying for each layer) and prohibited the use of physically based models. Here, it is demonstrated that an explicit calculation of infrared transmittances for each absorbing gas (H2O, CO2, O3, CH4, N2O, and O2) can be done efficiently, provided that a monochromatic approach is followed as in a regression model such as Radiative Transfer for TOVS (the TIROS Operational Vertical Sounder) (operational in most weather centers). The classical Goody random model is chosen as a physical formulation for spectral line absorption along with established water vapor and oxygen continua parameterizations. Line-by-line transmittance calculations for 189 atmospheric pr...

Both the issues of high-resolution satellite analysis and model evaluation for a region centered on the Arctic Circle (60°–75°N) are addressed. Model cloud fraction, cloud height, and outgoing radiation are compared to corresponding satellite observations using a model-to-satellite approach (calculated radiances from model state). The dataset consists of forecasts run at 15-km resolution up to 30 h and nearly coincident Advanced Very High Resolution Radiometer (AVHRR) imagery during the Beaufort and Arctic Storm Experiment over the Mackenzie Basin for a monthly period in the fall of 1994. A cloud detection algorithm is designed for day and night application using the 11-μ channel of AVHRR along with available information on atmospheric and ground temperatures. The satellite and model estimates of cloud fraction are also compared to observations at 20 ground stations. A significant result of the validation is that the model has a higher frequency of low cloud tops and a lower frequency of midlevel cloud tops than the observations. On a monthly basis, the model 11-μ outgoing brightness temperature (TB) is consequently higher than observed by about 4.4 K at all forecast times, which corresponds to a deficit of 760 m in mean cloud-top height and about 10 W m−2 in outgoing flux at the top of the atmosphere. Possible errors in the parameterization of ice or water cloud emissivity are evaluated but ruled out as the dominant cause for the warm TB bias in the model. Rather, the problem is attributed to low clouds being trapped in the boundary layer, whereas high clouds appear to be reasonably well modeled. The role of thin ice clouds is further evaluated by comparing distributions of observed and modeled 11-μ minus 12-μ TB differences, DIF45 (channel 4 minus channel 5). The relationship between the true height of the clouds and the effective height observed by satellite is modeled from forecast outputs as a function of DIF45. The quality of daily estimates is evaluated from time series at various locations. The time series shows that there was a marked drop in DIF45 during the month, which is attributed to a decrease in the occurrence of cirrus clouds. Finally, the diurnal cycle of TB and cloud fraction is found to be relatively large with average monthly 0600–1800 UTC TB differences of both signs of the order of 4–8 K in broad sectors and cloud fraction differences of 10%–30%. Where low clouds prevail, the cloud fraction tends to decrease at night and TB increases. Overall, model–observation differences are dominated by differences in the vertical distribution of clouds. A reduction of this effect implies a modification of the “preferred” model climatology in terms of its vertical distribution of humidity and cloud water.

A strong linearity exists between the 6.7-μm clear-sky outgoing brightness temperature (BT) and dewpoint depression (DPD) at upper-tropospheric levels. A similar relationship, using the logarithm of relative humidity instead of DPD, was developed by Soden and Bretherton. Here, however, the humidity at specific levels is derived as opposed to the humidity integrated over upper-tropospheric levels. Linear relationships are obtained between a 6-h model forecast of DPD and calculated BTs at different viewing angles. The data are further stratified in terms of 400-mb temperature as an indicator of airmass type. Applying these relationships using observed 6.7-μm BTs and a 6-h forecast of 400-mb temperature yields vertically correlated estimates of DPD between 200 and 500 mb, with DPD typically decreasing with height, and corresponding rms error estimates in the range 3–6 K. The retrieval technique is applied to GOES-8 and GOES-9 data, which cover about 40% of the globe. In cloudy regions, proxy humidit...

This study explores the feasibility of performing an objective analysis of instantaneous rain rate combining satellite estimates (and eventually other types of observations) with those from a numerical prediction model using the method of statistical interpolation. Results demonstrate that the quality of the short-term precipitation forecasts serving as background field has reached a level that makes such an objective analysis possible. The two main requirements to obtain an accurate analysis from available information are a realistic estimate of background field and observation errors and knowledge of the horizontal correlation of these errors with distance. The importance of specifying the errors for joint model-observation situations is emphasized; it is especially important in situations where model and observations are in conflict. These aspects of the problem are studied using collocated 6-h forecast with satellite estimates derived from visible and infrared imagery, and ground-truth rainfa...

Total precipitable water (TPW) retrieved from Special Sensor Microwave/lmager (SSM/I) brightness temperatures and specific humidity retrieved from Geostationary Operational Environmental Satellite (GOES) radiances are assimilated using a one-dimensional (ID) variational analysis technique. The study is divided into two parts. First, collocations with radiosondes are performed to arm the quality of the satellite water vapor retrievals. Collocations are also performed with 6-h forecast Acids. Second, SSM/I TPW and GOES specific humidity are assimilated using a ID variational analysis technique that minimizes the error variance of the analyzed field. A global collocation study over the oceans for SSM/I TPW retrievals and 6-h forecasts of TPW shows that the rmse (with respect to radiosondes) are, respectively, 4.7 and 5.0 kg m−2. A separate collocation study over both the oceans and land for GOES retrieved TPW and 6-h forecasts of TPW yields rmse of 4.6 and 4.4 kg m−2, respectively, in the midlatitudes and 6.8 and 5.9 kg m−2 in the Tropics. The reduction of the 6-h forecast rmse when assimilating SSM/I TPW is 1 kg m−2, which is a reduction of 20% in the rmse. When GOES retrievals of specific humidity are assimilated, the elective reduction is 0.6 kg m−2. It is shown that in the upper levels of the troposphere (above 600 mb), the error reduction of specific humidity is largely due to the GOES retrievals, whereas in the lower troposphere (850 and 700 mb), the reduction is mostly due to the SSM/I TPW. This emphasizes the complementarity of the information contained at different wavelengths and the advantage of using multisensor retrievals in data analysis.

Using Global Precipitation Climatology Project data gathered during June, July, and August 1989 over Japan, rainfall estimates are examined from both geostationary satellite imagery using a multifeature classification approach, and from short-term weather prediction model fields. Additionally, the utility of combining model forecast information within such a classifier to improve the final estimate is investigated. During both months satellite estimates are superior to model forecasts in detecting heavy rain events associated with extremely cold cloud tops, and in identifying cloud-free regions. Model estimates are superior to satellite retrievals in terms of dynamic range and regional bias. Addition of visible data to an infrared-only scheme improved monthly rainfall estimates during June, and hourly estimates during both months. In June it is shown that a combined satellite-model method clearly yields improved retrievals of rainfall relative to those obtained by using either satellite data or m...

A retrieval technique based on cloud classification is designed to derive humidity profiles from Meteosat visible (VIS), infrared window (IR), and water vapor (WV) channels, or equivalent sensors available on other satellites. Dewpoint depression (DPD) is the variable retrieved at six standard levels: 1000, 850, 700, 500, 400, and 300 mb. Collocation of soundings and Meteosal-2 imagery was obtained over Europe for March, June, and July 1988. Results are derived from over 2000 dependent and 1000 independent samples. It is found that a classification in seven (IR only) or nine (VIS-IR) classes contains the essential information on cloud type for the application sought. Measures were extracted from approximately 8-km pixel resolution images on 80-km × 80-km and 160-km × 160-km arm, little dependency on horizontal scale was found for the mean humidity profiles associated with each cloud class. The WV channel proved very useful in improving DPDs at higher levels while the VIS channel improved inferenc...

The usefulness of numerical weather prediction models in very short-range forecasting is limited by the spinup problem, resulting in an underestimation of both the divergent wind component and the precipitation. To alleviate the spinup problem, latent-heating profiles were directly assimilated into the Canadian regional finite-element (RFE) model. The estimates of latent heating were based on the precipitation rates inferred from GOES infrared and visible imagery. The latent heating was distributed in the vertical according to the stratiform condensation scheme of the model, but the heating rates were normalized to correspond to the satellite-inferred rain rates. The initial relative humidity field was enhanced to 95% between sigma-level 0.875 and the cloud top wherever the probability of precipitation, derived from satellite imagery, was larger than 40%. The results of a case study from the Canadian Atlantic Storms Program (CASP) indicated that the spinup time of the vertical motion, initially o...

Sunshine data are received daily on a real time basis at the Quebec Forecast Office and are converted into terms of percent of possible sunshine. On the basis of a survey held in the Quebec Forecast Office, the percent of possible sunshine was associated with the plain language expressions used to describe sky cover. Using this relationship, it was possible to use these observed sunshine reports to objectively verify the worded forecasts of sky cover. Techniques are being developed to forecast the percent of possible sunshine to be used as guidance in our public forecasts.

Infrared heating rate calculations with the emissivity formulation are improved by developing simple methods to minimize the effects of the two main sources of error 1) the use of the strong-line approximation; 2) the neglect of temperature variations along the optical path. When compared to band-by-hand calculations, the resulting new scheme maintains a 5% accuracy on cooling rates, and errors generally less than 2 W m−2 on fluxes. The most recent band parameters of MeClatchey et al. (1973) are used in the band-by-band calculations, including measured data for the vibration-rotation bands. Another novel aspect of the new scheme is the treatment of “e-type” absorption, which is a very important effect in hot humid air near the ground. An implicit parameterization is developed which allows for “e-type” absorption in the pre-tabulated transmission functions, and so requires no computational overhead. This is based on the derivation of an approximate formula which expresses the optical depth due to ...

A scheme is presented for the automated classification of oceanic cloud patterns. The 20 cloud classes reflect the rich variety of morphologies that are detectable from space. A training set is defined by 2000 samples of size 128 × 128 km taken from GOES visible and infrared images over the western Atlantic in February 1984. Class discrimination is obtained from 13 features representing height, albedo, shape and multilayering characteristics of the cloud fields. Two features derived from the two-dimensional power spectrum of the visible images proved essential for the detection of directional patterns (cloud “streets or rolls) and open cells. Based on the assumption of multinormal distributions of the features, a simple classification algorithm is developed. The generation of artificial samples yields a theoretical separability of 97% while the actual separability obtained on the training set is 95%. From 1020 independent samples, the separate verification of three expert nephanalysts indicates s...

The usefulness of cloud classification for detecting and quantifying air temperature and humidity anomalies above the ocean surface is examined. Cloud fields are classified in 20 classes following the automated method of Garand (1988), here applied over the northwestern Atlantic during the winter season. From collocation of the classified cloud fields (scale of ≈130 km) with ship or buoy observations of air temperature and humidity, significant anomalies are found for specific cloud classes while for other classes no anomaly is found. All results are verified from independent data taken in early 1984 and 1986. The results confirm that for the mesoscale cellular convective patterns (MCC), i.e., cloud “streets”, rolls, and open cells, the air and dew point temperatures are colder than climatology by several degrees, implying large latent and sensible heat fluxes. A latitudinal dependency of the anomaly is also observed. The removal of this bias provides estimates of surface air temperature with an ...

Two methods of deriving probability of precipitation fields (PP) over oceanic areas are presented and compared. The cloud fields are analyzed at the scale of ∼130–150 km from satellite visible and infrared imagery and collocated with ship observations of present weather. Method 1 is based on a detailed cloud classification scheme in 20 classes: a mean PP is determined for each cloud class. Method 2 assigns a PP based on cloud top temperature and mean cloud albedo only. For both methods, a normalization with respect to cloud fraction is applied. Method 1 involves more cloud field descriptors than Method 2, but the latter is simpler to implement and much faster. The PPs assigned to individual cloud fields vary between 0% and 65%. The importance of the visible sensor is clearly demonstrated, i.e., infrared-only techniques will be much less accurate. For real-time applications, the two methods provide similar results except for some specific cloud classes where maximum differences reach 13%, due to t...

A structural-stochastic image model is developed for the analysis and synthesis of cloud images. The ability of the model to characterize the visual appearance of cloud fields observed by satellite with a limited number of parameters is demonstrated. The model merges structural and stochastic information, the stochastic model acting as a local statistical operator applied to the output of the structural model. The structural or large-scale organization of the scene is retrieved from the two-dimensional Fourier representation of the digital image. The pattern generated by the major Fourier components provides a first guess of the scene. The stochastic aspect is described by a Markov model of texture that assumes a binomial probability distribution for the local grey-level variability. This Markov model provides four parameters that represent the clustering strength in the horizontal, vertical and diagonal directions. These parameters are estimated by a standard maximum-likelihood technique. The image can be reproduced with a fair degree of verisimilitude from these parameters. The data compression factor is of the order of one hundred to several hundreds.