Advanced Search
Article Contents

Application of an Artificial Neural Network for a Direct Estimation of Atmospheric Instability from a Next-Generation Imager


doi: 10.1007/s00376-015-5084-9

  • Atmospheric instability information derived from satellites plays an important role in short-term weather forecasting, especially the forecasting of severe convective storms. For the next generation of weather satellites for Korea's multi-purpose geostationary satellite program, a new imaging instrument has been developed. Although this imaging instrument is not designed to perform full sounding missions and its capability is limited, its multi-spectral infrared channels provide information on vertical sounding. To take full advantage of the observation data from the much improved spatiotemporal resolution of the imager, the feasibility of an artificial neural network approach for the derivation of the atmospheric instability is investigated. The multi-layer perceptron model with a feed-forward and back-propagation training algorithm shows quite a sensitive response to the selection of the training dataset and model architecture. Through an extensive performance test with a carefully selected training dataset of 7197 independent profiles, the model architectures are selected to be 12, 5000, and 0.3 for the number of hidden nodes, number of epochs, and learning rate, respectively. The selected model gives a mean absolute error, RMSE, and correlation coefficient of 330 J kg-1, 420 J kg-1, and 0.9, respectively. The feasibility is further demonstrated via application of the model to real observation data from a similar instrument that has comparable observation channels with the planned imager.
  • 加载中
  • Berk A., G. P. Anderson, P. K. Acharya, and E. P. Shettle, 2011: MODTRAN\textregistered 5.2.2 User-檚 Manual. Spectral Sciences, INC., Burlington, MA, 69 pp.
    Blackwell W. J., F. W. Chen, 2009: Introduction to multilayer perceptron neural networks. Neural Networks in Atmospheric Remote Sensing, Massachusetts Institute of Technology, 73- 96.10.1080/01431169721870067472c760e0c0d41e3a07f3aaffac5adhttp%3A%2F%2Fwww.tandfonline.com%2Fdoi%2Fpdf%2F10.1080%2F014311697218700http://www.tandfonline.com/doi/pdf/10.1080/014311697218700ABSTRACT Over the past decade there have been considerable increases in both the quantity of remotely sensed data available and the use of neural networks. These increases have largely taken place in parallel, and it is only recently that several researchers have begun to apply neural networks to remotely sensed data. This paper introduces this special issue which is concerned specifically with the use of neural networks in remote sensing. The feed-forward back-propagation multi-layer perceptron (MLP) is the type of neural network most commonly encountered in remote sensing and is used in many of the papers in this special issue. The basic structure of the MLP algorithm is described in some detail while some other types of neural network are mentioned. The most common applications of neural networks in remote sensing are considered, particularly those concerned with the classification of land and clouds, and recent developments in these areas are described. Finally, the application of neural networks to multi-source data and fuzzy classification are considered.
    Botes D., J. R. Mecikalski, and G. J. Jedlovec, 2012: Atmospheric Infrared Sounder (AIRS) sounding evaluation and analysis of the pre-convective environment. J. Geophys. Res., 117(D9), doi: 10.1029/2011JD016996.10.1029/2011JD0169969dd31fca601e04335441d70faac82b2chttp%3A%2F%2Fonlinelibrary.wiley.com%2Fdoi%2F10.1029%2F2011JD016996%2Fpdfhttp://onlinelibrary.wiley.com/doi/10.1029/2011JD016996/pdf[1] The Atmospheric Infrared Sounder (AIRS) is a hyperspectral instrument onboard the National Aeronautics and Space Administration's (NASA) Earth Observing System (EOS) Aqua satellite. This study investigates the performance of AIRS soundings in characterizing the stability in the pre-convective environment of the southeastern United States. AIRS soundings are collocated with radiosonde observations within 卤1 degree and 2 h of the Aqua overpass. For each case, the AIRS sounding with maximum PBest quality indicator (signifying the pressure level above which the sounding is of best quality) is chosen for analysis. Rapid Update Cycle soundings from 1800 UTC analyses are used to evaluate the results from AIRS. Precipitable water and stability indices including convective available potential energy, convective inhibition, Lifted Index, K-Index, and Total Totals are derived from all soundings. Results indicate that AIRS underestimates instability due to a dry bias at the surface and roughly 900 hPa. A simple method is presented for reconstructing a RAOB-like inversion (in terms of magnitude and altitude) within AIRS soundings, hence developing more representative RAOB-like soundings that can benefit the operational forecaster.
    Craven J. P., R. E. Jewell, and H. E. Brooks, 2002: Comparison between observed convective cloud-base heights and lifting condensation level for two different lifted parcels. Wea. Forecasting, 17, 885- 890.
    EUMETSAT, 2013: ATBD for the MSG GII/TOZ product. EUM/ MET/DOC/11/0247,32 pp.
    Gardner M. W., S. R. Dorling, 1998: Artificial neural networks (the multilayer perceptron): A review of applications in the atmospheric sciences. Atmos. Environ., 32, 2627- 2636.10.1016/S1352-2310(97)00447-0ed69df7e-c2e9-44cd-92e3-663cee2120a9d28230ad05f45455c650b806a5796e42http%3A%2F%2Fwww.researchgate.net%2Fpublication%2F263416087_Artificial_neural_network_%28Multilayer_Perceptron%29A_review_of_applications_in_atmospheric_sciencesrefpaperuri:(72690f7281eb718f0276a81000ee067f)http://www.researchgate.net/publication/263416087_Artificial_neural_network_(Multilayer_Perceptron)A_review_of_applications_in_atmospheric_sciencesABSTRACT Artificial neural networks are appearing as useful alternatives to traditional statistical modelling techniques in many scientific disciplines. This paper presents a general introduction and discussion of recent applications of the multilayer perceptron, one type of artificial neural network, in the atmospheric sciences.
    Hilton F., Coauthors, 2012: Hyperspectral earth observation from IASI: Five years of accomplishments. Bull. Amer. Meteor. Soc., 93( 3), 347- 370.10.1175/BAMS-D-11-00027.17d1a1845baffef16ed54a5f1105aecbehttp%3A%2F%2Fwww.researchgate.net%2Fpublication%2F231315606_Hyperspectral_Earth_Observation_from_IASI_Five_Years_of_Accomplishmentshttp://www.researchgate.net/publication/231315606_Hyperspectral_Earth_Observation_from_IASI_Five_Years_of_AccomplishmentsThe Infrared Atmospheric Sounding Interferometer (IASI) forms the main infrared sounding component of the European Organisation for the Exploitation of Meteorological Satellites's (EUMETSAT's) Meteorological Operation (MetOp)-A satellite (Klaes et al. 2007), which was launched in October 2006. This article presents the results of the first 4 yr of the operational IASI mission. The performance of the instrument is shown to be exceptional in terms of calibration and stability. The quality of the data has allowed the rapid use of the observations in operational numerical weather prediction (NWP) and the development of new products for atmospheric chemistry and climate studies, some of which were unexpected before launch. The assimilation of IASI observations in NWP models provides a significant forecast impact; in most cases the impact has been shown to be at least as large as for any previous instrument. In atmospheric chemistry, global distributions of gases, such as ozone and carbon monoxide, can be produced in near-搑eal time, and short-lived species, such as ammonia or methanol, can be mapped, allowing the identification of new sources. The data have also shown the ability to track the location and chemistry of gaseous plumes and particles associated with volcanic eruptions and fires, providing valuable data for air quality monitoring and aircraft safety. IASI also contributes to the establishment of robust long-term data records of several essential climate variables. The suite of products being developed from IASI continues to expand as the data are investigated, and further impacts are expected from increased use of the data in NWP and climate studies in the coming years. The instrument has set a high standard for future operational hyperspectral infrared sounders and has demonstrated that such instruments have a vital role in the global observing system.
    Jin X., J. Li, 2010: Improving moisture profile retrieval from broadband infrared radiances with an optimized first-guess scheme. Remote Sensing Letters ,1(4), 231-238, doi:10.1080/01431161003762322.10.1080/01431161003762322c49a1f2c7a3e8afcd742d7c47d5f9022http%3A%2F%2Fwww.tandfonline.com%2Fdoi%2Fabs%2F10.1080%2F01431161003762322http://www.tandfonline.com/doi/abs/10.1080/01431161003762322Variational retrieval of legacy atmospheric moisture profiles needs to begin with a first guess. An optimized first-guess scheme is developed for moisture profile retrieval from broadband infrared (IR) radiances. In this scheme, the non-exponential response of moisturemixingratio toIRradianceat high temperatures (>273 K) is considered. It is found that the first guess of low-level (below 550 hPa) moisture profiles is substantially improved after the new scheme. The data collected by Spinning Enhanced Visible and Infrared Imager (SEVIRI) onboard the Meteosat Second Generation are used for validation. This scheme provides an important optimization method for the next generation of Geostationary Operational Environmental Satellite (GOES)-R legacy profile retrieval algorithm because the Advanced Baseline Imager (ABI) onboard the GOES-R has very similar configurations to SEVIRI.
    Jin X., J. Li, T. J. Schmit, J. L. Li, M. D. Goldberg, and J. J. Gurka, 2008: Retrieving clear-sky atmospheric parameters from SEVIRI and ABI infrared radiances. J. Geophys. Res., 113,D15310, doi: 10.1029/2008JD010040.10.1029/2008JD01004071a69a2c-105e-4cfc-8604-1e296bf9d421ede66aed5c58fddbe049033b7204ee6ehttp%3A%2F%2Fonlinelibrary.wiley.com%2Fdoi%2F10.1029%2F2008JD010040%2Fpdfrefpaperuri:(2aa7302f0ff2d34c9c88361fd2f78b86)http://onlinelibrary.wiley.com/doi/10.1029/2008JD010040/pdf[1] The algorithm for the current Geostationary Operational Environmental Satellite (GOES) Sounders is adapted to produce atmospheric temperature and moisture legacy profiles from simulated infrared radiances of the Advanced Baseline Imager (ABI) on board the next generation GOES-R. Since the Spinning Enhanced Visible and Infrared Imager (SEVIRI) on board the Meteosat Second Generation (MSG) Meteosat-8/9 has many of the same spectral and spatial features as ABI, it is used as proxy to test the algorithm. Because as imagers, SEVIRI and ABI do not have enough CO 2 absorption spectral bands relative to the current GOES Sounders, the legacy profile algorithm for the current GOES Sounders needs to be modified. Both simulations and analysis with radiance measurements indicate that the single temperature-sensitive infrared band (13.4 m) of SEVIRI cannot provide enough temperature profile information. However, SEVIRI's two H 2 O absorption spectral bands (6.2 and 7.2 m) are able to provide useful information on water vapor content above 700 hPa. Because of their high spatial (approximately 3 km for SEVIRI and 2 km for ABI IR bands) and high temporal (15 min full disk coverage) resolutions, SEVIRI and ABI will provide useful profile products with a quality similar to that from the current GOES Sounder prior to the availability of a hyperspectral IR sounding system in geostationary orbit.
    Kitzmiller D. H., W. E. McGovern, 1989: VAS retrievals as a source of information for convective weather forecasts: An objective assessment and comparison with other sources of upper-air observations. Mon. Wea. Rev., 117, 2095- 2110.
    Kim D. H., M. H. Ahn, 2014: Introduction of the in-orbit test and its performance for the first meteorological imager of the Communication, Ocean, and Meteorological Satellite. Atmospheric Measurement Techniques, 7, 2471- 2485.10.5194/amt-7-2471-2014f9f41091607bf5acbf8badf38a1aebbchttp%3A%2F%2Fwww.researchgate.net%2Fpublication%2F266322147_Introduction_of_in_orbit_test_and_its_performance_of_the_first_meteorological_imager_of_the_Communication_Ocean_and_Meteorological_Satellitehttp://www.researchgate.net/publication/266322147_Introduction_of_in_orbit_test_and_its_performance_of_the_first_meteorological_imager_of_the_Communication_Ocean_and_Meteorological_SatelliteThe first geostationary Earth observation satellite of Korea - the Communication, Ocean, and Meteorological Satellite (COMS) - was successfully launched on 27 June 2010. After arrival at its operational orbit, the satellite underwent an in-orbit test (IOT) that lasted for about 8 months. During the IOT period, the main payload for the weather application, the meteorological imager, went through successful tests for demonstrating its function and performance, and the test results are introduced here. The radiometric performance of the meteorological imager (MI) is tested by means of signal-to-noise ratio (SNR) for the visible channel, noise-equivalent differential temperature (NEdT) for the infrared channels, and pixel-to-pixel nonuniformity for both the visible and infrared channels. In the case of the visible channel, the SNR of all eight detectors is obtained using the ground-measured parameters with the background signals obtained in orbit. The overall performance shows a value larger than 26 at 5% albedo, exceeding the user requirement of 10 by a significant margin. Also, the relative variability of detector responsivity among the eight visible channels meets the user requirement, showing values within 10% of the user requirement. For the infrared channels, the NEdT of each detector is well within the user requirement and is comparable with or better than the legacy instruments, except for the water vapor channel, which is slightly noisier than the legacy instruments. The variability of detector responsivity of infrared channels is also below the user requirement, within 40% of the requirement, except for the shortwave infrared channel. The improved performance result is partly due to the stable and low detector temperature obtained due to spacecraft design, i.e., by installing a single solar panel on the opposite side of the MI.
    Koenig M., E. de Coning, 2009: The MSG global instability indices product and its use as a nowcasting tool. Wea. Forecasting, 24, 272- 282.10.1175/2008WAF2222141.1b6bd9c3a-95bf-4858-b4e6-43a73351d3596d15d8a9930c2c5d6b87ec571521f0d0http%3A%2F%2Fwww.researchgate.net%2Fpublication%2F240687258_The_MSG_Global_Instability_Indices_Product_and_Its_Use_as_a_Nowcasting_Toolrefpaperuri:(b87e65f3f46f346a0399bf19da1c2d4f)http://www.researchgate.net/publication/240687258_The_MSG_Global_Instability_Indices_Product_and_Its_Use_as_a_Nowcasting_ToolThe European geostationary Meteosat Second Generation (MSG) satellite offers a variety of channels to use for various purposes, including nowcasting of convection. A number of applications have also been developed to make use of these new capabilities for nowcasting, especially for the detection and prediction of severe weather. The MSG infrared channel selection makes it possible to assess the air stability in preconvective, that is, still cloud-free, conditions. Instability indices are traditionally derived from radiosonde profiles. Such indices typically combine measures of the thermal and moisture properties and often only use a small quantity of vertical profile parameters. MSG-based temperature and moisture retrievals are used for the derivation of stability indices, which are a part of the MSG meteorological products derived centrally at the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT). Such indices are of an empirical nature, are often only applicable to certain geographic regions, and their thresholds are dependent on seasonal variation, but they can assess the likelihood of convection within the next few hours, thus providing a warning lead of about 6芒鈧-9 h. Numerous test cases and the more quantitative verification process that was initiated by the South African Weather Service show the generally good warning potential of the derived instability fields. The added capability of a nearly continuous monitoring of the instability fields that is guaranteed by MSG's 15-min repeat cycle is most valuable, since it provides nowcasters with new information much more regularly than the twice-a-day soundings at only a limited number of radiosonde stations. The current EUMETSAT instability product is aimed at helping forecasters to focus their attention on a certain region, which can then be monitored more closely with other means, like satellite imagery and radar data, over the next few hours.
    Krasnopolsky V. M., 2007: Neural network emulations for complex multidimensional geophysical mappings: Applications of neural network techniques to atmospheric and oceanic satellite retrievals and numerical modeling. Rev. Geophys., 45,RG3009, doi: 10.1029/2006RG000200.10.1029/2006RG000200dd2025617ccc3c612a6faa6c2b9a8ceehttp%3A%2F%2Fonlinelibrary.wiley.com%2Fdoi%2F10.1029%2F2006RG000200%2Fpdfhttp://onlinelibrary.wiley.com/doi/10.1029/2006RG000200/pdf[1] A group of geophysical applications, which from the mathematical point of view, can be formulated as complex, multidimensional, nonlinear mappings and which in terms of the neural network (NN) technique, utilize a particular type of NN, the multilayer perceptron (MLP), is reviewed in this paper. This type of NN application covers the majority of NN applications developed in geosciences like satellite remote sensing, meteorology, oceanography, numerical weather prediction, and climate studies. The major properties of the mappings and MLP NNs are formulated and discussed. Three particular groups of NN applications are presented in this paper as illustrations: atmospheric and oceanic satellite remote sensing applications, NN emulations of model physics for developing atmospheric and oceanic hybrid numerical models, and NN emulations of the dependencies between model variables for application in data assimilation systems.
    Lee S. J., M. H. Ahn, and Y. Lee, 2013: Application of artificial neural network for direct estimation of atmospheric instability Index (CAPE) from geostationary satellite. Proceedings of Autumn Meeting of KMS, 544-545, Gwang-ju, Korea, Kor. Meteo. Soc.
    Lee S. J., M. H. Ahn, and Y. Lee, 2014a: Application of artificial neural network for the direct estimation of atmospheric instability from a geostationary satellite imager. 10pp, Proceedings of the 19th ITSC, Jeju Island, South Korea.
    Lee Y. -K., Z. L. Li, J. Li, and T. J. Schmit, 2014b: Evaluation of the GOES-R ABI LAP retrieval algorithm using the GOES-13 sounder. J. Atmos. Oceanic Technol.,31, 3-19, doi: 10.1175/JTECH-D-13-00028.1.10.1175/JTECH-D-13-00028.1a85bf888963c7c80e1a427c4dd9cd7adhttp%3A%2F%2Fwww.researchgate.net%2Fpublication%2F274493065_Evaluation_of_the_GOES-R_ABI_LAP_Retrieval_Algorithm_Using_the_GOES-13_Sounderhttp://www.researchgate.net/publication/274493065_Evaluation_of_the_GOES-R_ABI_LAP_Retrieval_Algorithm_Using_the_GOES-13_SounderABSTRACT A physical retrieval algorithm has been developed for deriving the legacy atmospheric profile (LAP) product from infrared radiances of the Advanced Baseline Imager (ABI) on board the next-generation Geostationary Operational Environmental Satellite (GOES-R) series. In this study, the GOES-R ABI LAP retrieval algorithm is applied to the GOES-13 sounder radiance measurements (termed the GOES-13 LAP retrieval algorithm in this study) for its validation as well as for potential transition of the GOES-13 LAP retrieval algorithm for the operational processing of GOES sounder data. The GOES-13 LAP retrievals are compared with five different truth measurements: radiosonde observation (raob) and microwave radiometer-measured total precipitable water (TPW) at the Atmospheric Radiation Measurement Cloud and Radiation Testbed site, conventional raob, TPW measurements from the global positioning system-integrated precipitable water NOAA network, and TPW measurements from the Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E). The results show that with the GOES-R ABI LAP retrieval algorithm, the GOES-13 sounder provides better water vapor profiles than the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) forecast fields at the levels between 300 and 700 hPa. The root-mean-square error (RMSE) and standard deviation (STD) of the GOES-13 sounder TPW are consistently reduced from those of the GFS forecast no matter which measurements are used as the truth. These substantial improvements indicate that the GOES-R ABI LAP retrieval algorithm is well prepared to provide continuity of quality to some of the current GOES sounder products, and the algorithm can be transferred to process the current GOES sounder measurements for operational product generation.
    Li J., C. -Y. Liu, P. Zhang, and T. J. Schmit, 2012: Applications of full spatial resolution space-based advanced infrared soundings in the preconvection environment. Wea. Forecasting, 27, 515- 524.10.1175/WAF-D-10-05057.1d98866a4-497b-4a50-a188-0b008cc632383605c939c8bf3cc3796c857bfb02fc6chttp%3A%2F%2Fwww.researchgate.net%2Fpublication%2F258724591_Applications_of_Full_Spatial_Resolution_Space-Based_Advanced_Infrared_Soundings_in_the_Preconvection_Environmentrefpaperuri:(1bfedd0ba8665571a7f03e0d948da085)http://www.researchgate.net/publication/258724591_Applications_of_Full_Spatial_Resolution_Space-Based_Advanced_Infrared_Soundings_in_the_Preconvection_EnvironmentAbstract Advanced infrared (IR) sounders such as the Atmospheric Infrared Sounder (AIRS) and Infrared Atmospheric Sounding Interferometer (IASI) provide atmospheric temperature and moisture profiles with high vertical resolution and high accuracy in preconvection environments. The derived atmospheric stability indices such as convective available potential energy (CAPE) and lifted index (LI) from advanced IR soundings can provide critical information 1 ~ 6 h before the development of severe convective storms. Three convective storms are selected for the evaluation of applying AIRS full spatial resolution soundings and the derived products on providing warning information in the preconvection environments. In the first case, the AIRS full spatial resolution soundings revealed local extremely high atmospheric instability 3 h ahead of the convection on the leading edge of a frontal system, while the second case demonstrates that the extremely high atmospheric instability is associated with the local development of severe thunderstorm in the following hours. The third case is a local severe storm that occurred on 7-8 August 2010 in Zhou Qu, China, which caused more than 1400 deaths and left another 300 or more people missing. The AIRS full spatial resolution LI product shows the atmospheric instability 3.5 h before the storm genesis. The CAPE and LI from AIRS full spatial resolution and operational AIRS/AMSU soundings along with Geostationary Operational Environmental Satellite (GOES) Sounder derived product image (DPI) products were analyzed and compared. Case studies show that full spatial resolution AIRS retrievals provide more useful warning information in the preconvection environments for determining favorable locations for convective initiation (CI) than do the coarser spatial resolution operational soundings and lower spectral resolution GOES Sounder retrievals.
    Li Z. L., J. Li, J. W. P. Menzel, T. J. Schmit, J. P. Nelson III, J. Daniels, and S. A. Ackerman, 2008: GOES sounding improvement and applications to severe storm nowcasting. Geophys. Res. Lett., 35,L03806, doi: 10.1029/2007GL032797.10.1029/2007GL0327979bec177b29006bd198403af69eb10396http%3A%2F%2Fonlinelibrary.wiley.com%2Fdoi%2F10.1029%2F2007GL032797%2Ffullhttp://onlinelibrary.wiley.com/doi/10.1029/2007GL032797/full[1] An improved clear-sky physical retrieval algorithm for atmospheric temperature and moisture is applied to the Geostationary Operational Environmental Satellite-12 (GOES-12) Sounder. A comparison with the microwave radiometer (MWR) measured total precipitable water (TPW) at the Southern Great Plains (SGP) Cloud and Radiation Testbed (CART) site from June 2003 to May 2005 shows that the TPW retrievals are improved by 0.4 mm over the legacy GOES Sounder TPW product. The Lifted Index (LI) derived product imagery (DPI) from the improved soundings better depicts the pre-convective environment surrounding a tornadic supercell at Eagle Pass, Texas on 24 April 2007. Another severe storm case from 13 April 2006 demonstrates that the improved physical algorithm successfully detects low-level moisture. Both cases show the new retrievals along with the derived products will help the forecasters with short-term severe storm nowcasting.
    Liu H., C. Wu, J. Li, and Q. Chengli, 2014: Deriving atmospheric instability indices directly from Geostationary Interferometric Infrared Sounder (GIIRS) radiances. poster presentation in the 19th ITSC, Jeju Island, South Korea.
    Ma X. L., T. J. Schmit, and W. L. Smith, 1999: A nonlinear physical retrieval algorithm-its application to the GOES-8/9 sounder. J. Appl. Meteor., 38, 501- 513.10.1175/1520-0450(1999)038<0501:ANPRAI>2.0.CO;2712d4ce2-d99c-4806-8716-3aa8de73186b6a5d5cb4e0983766db17c63e9608fa60http%3A%2F%2Fwww.researchgate.net%2Fpublication%2F252954623_A_Nonlinear_Physical_Retrieval_Algorithm--Its_Application_to_the_GOES89_Sounderrefpaperuri:(6fba49d314dc3999f93e66a2bae7af3a)http://www.researchgate.net/publication/252954623_A_Nonlinear_Physical_Retrieval_Algorithm--Its_Application_to_the_GOES89_SounderAbstract A nonlinear physical retrieval algorithm is developed and applied to the GOES-8/9 sounder radiance observations. The algorithm utilizes Newtonian iteration in which the maximum probability solution for temperature and water vapor profiles is achieved through the inverse solution of the nonlinear radiative transfer equation. The nonlinear physical retrieval algorithm has been tested for one year. It has also been implemented operationally by the National Oceanic and Atmospheric Administration National Environmental Satellite, Data and Information Service during February 1997. Results show that the GOES retrievals of temperature and moisture obtained with the nonlinear algorithm more closely agree with collocated radiosondes than the National Centers for Environmental Prediction (NCEP) forecast temperature and moisture profile used as the initial profile for the solution. The root-mean-square error of the total water vapor from the solution first guess, which is the NCEP 12-h forecast (referred to as the -渂ackground-), is reduced approximately 20% over the conventional data-rich North American region with the largest changes being achieved in areas of sparse radiosonde data coverage.
    Martinez M. A., M. Velazquez, M. Manso, and I. Mas, 2007: Application of LPW and SAI SAFNWC/MSG satellite products in pre-convective environments. Atmospheric Research, 83, 366- 379.10.1016/j.atmosres.2005.10.0222fdc0d15-4e27-446a-b39c-cff629bebf773218f45418634224f668397adba075dbhttp%3A%2F%2Fwww.sciencedirect.com%2Fscience%2Farticle%2Fpii%2FS0169809506001347refpaperuri:(e40c30320ab4c5ea21e8fdea6a035544)http://www.sciencedirect.com/science/article/pii/S0169809506001347ABSTRACT The Spinning Enhanced Visible and Infrared Imagery (SEVIRI) instrument, on board the Meteosat Second Generation (MSG), is a radiometer with eight infrared (IR) spectral bands. Seven of these channels are used to retrieve Layer Precipitable Water (LPW) and Stability Analysis Imagery (SAI). Both products are the PGE07 and the PGE08 of SAFNWC (Satellite Application Facility on support to Nowcasting and Very Short-Range Forecasting). The authors at Instituto Nacional de Meteorolog&iacute;a (INM) have developed the LPW and SAI algorithms, in the SAFNWC framework. Both products are retrieved using statistical retrieval based on neural networks. The main advantage of these algorithms versus physical retrieval algorithms is the independence from the Numerical Weather Prediction (NWP) models. The LPW provides information on the water vapor contained in a vertical column of unit cross-section area in three layers in the troposphere (low, middle and high) and in the total layer in cloud free areas. The SAI provides estimations of the atmospheric instability in cloud free areas, in particular the Lifted Index (LI).The stability and precipitable water obtained with both products are routinely generated every 15 min at a satellite horizontal resolution of 3 km in NADIR. A significant advantage of these MSG products, compared to traditional measurements such as radiosondes, is their ability to measure high resolution temporal and spatial variations of atmospheric stability and moisture in pre-convective environments. The main disadvantage is that they do not have the vertical resolution of radiosonde. The MSG moisture and stability time trend fields are especially useful during the period preceding the outbreak of convection due to the high resolution. Once the outbreak of convection occurs, the products calculated in the clear air pixels surrounding the convective system will allow to foresee the evolution of the convection.
    Menzel W. P., F. C. Holt, T. J. Schmit, R. M. Aune, A. J. Schreiner, G. S. Wade, and D. G. Gray, 1998: Application of GOES-8/9 soundings to weather forecasting and nowcasting. Bull. Amer. Meteor. Soc., 79( 10), 2059- 2077.10.1175/1520-0477(1998)0792.0.CO;22d7924a18ee1af33c05901c9a94e7b64http%3A%2F%2Fwww.researchgate.net%2Fpublication%2F249615689_Application_of_GOES89_Soundings_to_Weather_Forecasting_and_Nowcastinghttp://www.researchgate.net/publication/249615689_Application_of_GOES89_Soundings_to_Weather_Forecasting_and_NowcastingCompares an updated set of geostationary sounders which were used to measure atmospheric radiances in 18 infrared spectral bands since 1994, to those available from the generation of Geostationary Operational Environmental Satellites (GOES). Examination of the anticipated movements from the GOES-8/9 sounders; Effectiveness of sounding projects over North America using the GOES-8/9 sounders.
    Oolman L., 2014: Upper Air Data Soundings. University of Wyoming, College of Engineering, Department of Atmospheric Science . [Available online at http://weather.uwyo.edu/],accessed in July 2014.10.1109/MCSE.2012.71bd6daf7998bce17094ca6fbd439ef824http%3A%2F%2Fetc.usf.edu%2Fclipart%2F44700%2F44764%2F44764_uni_wy.htmhttp://etc.usf.edu/clipart/44700/44764/44764_uni_wy.htmWe met with Provost Al Karnig a month ago and expressed concern about how the review process works. We do not think it works well. Administrators should be hired to carry out the policy of the board of trustees. It appears that the faculty and staff of the University have
    Romero R., M. Gay, and C. A. Dowsell III, 2007: European climatology of severe convective storm environmental parameters: a test for significant tornado events. Atmospheric Research, 83, 389- 404.10.1016/j.atmosres.2005.06.011ec1dd2abe1b5a0e21b5a131d9eb66dd6http%3A%2F%2Fwww.sciencedirect.com%2Fscience%2Farticle%2Fpii%2FS0169809506001360http://www.sciencedirect.com/science/article/pii/S0169809506001360ABSTRACT A climatology of various parameters associated with severe convective storms has been constructed for Europe. This involves using the reanalysis data base from ERA-40 for the period 1971-2000 and calculating monthly means, variability range and extremes occurrence of fields such as convective available potential energy, convective inhibition energy, mid-tropospheric lapse rate, low-tropospheric moisture content and storm relative helicity for different layers. This process is a first step towards development of a synthetic climatology of European severe weather, and is publicly available at the web site http://ecss.uib.es. Preliminary results derived from these products were presented during the ECSS 2004 conference. This paper is devoted to a more detailed presentation and discussion of the main results. It is hypothesized that preferred areas for severe thunderstorms occurrence in Europe would extend along a zonal belt over the south-central regions, where high helicity associated with the extratropical storm tracks and thermodynamically-favourable profiles established over the southern Atlantic and Mediterranean Sea would most likely be concatenated.Further, this effort has been complemented with a collection of existing reports of significant (at least F2) tornadoes in Europe during the period 1971-2003. We present this data set in this paper and it also can be found at the website http://ecss.uib.es. Thus, the tornado collection can be used to test the appropriateness of the parameters selected for the synthetic climatology. In particular, it is found that the convective available potential energy, low-tropospheric moisture content and environmental shear, when related to the monthly climatology, are reasonably good descriptors of the tornadic environments.
    Schmit T. J., J. Li, J. L. Li, W. F. Feltz, J. J. Gurka, M. D. Goldberg, and K. J. Schrab, 2008: The GOES-R advanced baseline imager and the continuation of current sounder products. Journal of Applied Meteorology and Climatology,47, 2696-2711, doi: 10.1175/2008JAMC1858.1.10.1175/2008JAMC1858.122eb93e1bf2b562d707359a4be805d42http%3A%2F%2Fwww.researchgate.net%2Fpublication%2F238428465_The_GOES-R_Advanced_Baseline_Imager_and_the_Continuation_of_Current_Sounder_Productshttp://www.researchgate.net/publication/238428465_The_GOES-R_Advanced_Baseline_Imager_and_the_Continuation_of_Current_Sounder_ProductsAbstract The first of the next-generation series of Geostationary Operational Environmental Satellites (GOES-R) is scheduled for launch in the 2015 time frame. One of the primary instruments on GOES-R, the Advanced Baseline Imager (ABI), will offer more spectral bands, higher spatial resolution, and faster imaging than does the current GOES Imager. Measurements from the ABI will be used for a wide range of qualitative and quantitative weather, land, ocean, cryosphere, environmental, and climate applications. However, the first and, likely, the second of the new series of GOES will not carry an infrared sounder dedicated to acquiring high-vertical-resolution atmospheric temperature and humidity profiles that are key to mesoscale and regional severe-weather forecasting. The ABI will provide some continuity of the current sounder products to bridge the gap until the advent of the GOES advanced infrared sounder. Both theoretical analysis and retrieval simulations show that data from the ABI can be combined with temperature and moisture information from forecast models to produce derived products that will be adequate substitutes for the legacy products from the current GOES sounders. Products generated from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) measurements also demonstrate the utility of those legacy products for nowcasting applications. However, because of very coarse vertical resolution and limited accuracy in the legacy sounding products, placing a hyperspectral-resolution infrared sounder with high temporal resolution on future GOES is an essential step toward realizing substantial improvements in mesoscale and severe-weather forecasting required by the user communities.
    Seemann S. W., J. Li, W. P. Menzel, and L. E. Gumley, 2003: Operational retrieval of atmospheric temperature, moisture, and ozone from MODIS infrared radiances. J. Appl. Meteor., 42, 1072- 1091.10.1117/12.4666860ae777e6-12ba-41d3-be16-f7ed5e73871d025d3731fe69135e53ba908d5fa6d477http%3A%2F%2Fwww.researchgate.net%2Fpublication%2F228942078_Operational_Retrieval_of_Atmospheric_Temperature_Moisture_and_Ozone_from_MODIS_Infrared_Radiancesrefpaperuri:(6eb821970c67af751e14dabfadc4cba1)http://www.researchgate.net/publication/228942078_Operational_Retrieval_of_Atmospheric_Temperature_Moisture_and_Ozone_from_MODIS_Infrared_RadiancesThe algorithm for retrieving atmospheric temperature, moisture, and total column ozone using the Moderate Resolution Imaging Spectroradiometer (MODIS) longwave infrared radiances is presented. The operational MODIS algorithm performs clear sky retrievals globally over land and ocean for both day and night. The algorithm is based on a regression and has an option to follow the statistical retrieval with a nonlinear physical retrieval. The regression coefficients are determined from an extension of the NOAA-88 data set containing more than 8400 global radiosonde measurements of atmospheric temperature, moisture and ozone profiles. Evaluation of atmospheric products is performed by a comparison with data from ground-based instrumentation, geostationary infrared sounders, and polar orbiting microwave sounders. MODIS moisture products are in general agreement with the gradients and distributions from the other satellites, while MODIS depicts more detailed structure with its improved spatial resolution.
    Seidel D. J., B. Sun, M. Pettey, and A. Reale, 2011: Global radiosonde balloon drift statistics. J. Geophys. Res. , 116,D07102, doi:10.1029/2010JD014891.10.1029/2010JD014891709efe04cfad2b51a68ef8d540578254http%3A%2F%2Fonlinelibrary.wiley.com%2Fdoi%2F10.1029%2F2010JD014891%2Fpdfhttp://onlinelibrary.wiley.com/doi/10.1029/2010JD014891/pdf[1] The drift of radiosonde balloons during their ascent has generally been considered a negligible factor in applications involving radiosonde observations. However, several applications envisioned for observations from the Global Climate Observing System (GCOS) Reference Upper Air Network (GRUAN) require estimates of balloon drift. This study presents a comprehensive global climatology of radiosonde balloon drift distance and ascent time, based on 2 years of data from 419 stations, with particular attention to GRUAN stations. Typical drift distances are a few kilometers in the lower troposphere, 655 km in the midtroposphere, 6520 km in the upper troposphere, and 6550 km in the lower stratosphere, although there is considerable variability due to variability in climatological winds. Drift distances tend to increase with height above the surface, be larger in midlatitudes than in the tropics, be larger in winter than in summer, and vary with wind (and consequent balloon drift) direction. Most estimates of elapsed time from balloon launch to various pressure levels, due to vertical balloon rise, have median values ranging from about 5 min at 850 hPa to about 1.7 h at 10 hPa, with ranges of about 20% of median values. Observed elapsed times exceed those estimated using assumed 5 or 6 m/s rise rates.
    Setv谩k, M., J. M眉ller, 2013: 2.5-minute rapid scan experiments with the MSG satellites. 7th European Conference on Severe Storms, Helsinki, Finland, 3-7June 2013.
    Taravat A., S. Proud, S. Peronaci, F. de Frate, and N. Oppelt, 2015: Multilayer perceptron neural networks model for meteosat second generation SEVIRI daytime cloud masking. Remote Sensing, 7, 1529- 1539.10.3390/rs702015291bb87f74125741a172ef52ffe5154b8dhttp%3A%2F%2Fwww.researchgate.net%2Fpublication%2F273280497_Multilayer_Perceptron_Neural_Networks_Model_for_Meteosat_Second_Generation_SEVIRI_Daytime_Cloud_Maskinghttp://www.researchgate.net/publication/273280497_Multilayer_Perceptron_Neural_Networks_Model_for_Meteosat_Second_Generation_SEVIRI_Daytime_Cloud_MaskingA multilayer perceptron neural network cloud mask for Meteosat Second Generation SEVIRI (Spinning Enhanced Visible and Infrared Imager) images is introduced and evaluated. The model is trained for cloud detection on MSG SEVIRI daytime data. It consists of a multi-layer perceptron with one hidden sigmoid layer, trained with the error back-propagation algorithm. The model is fed by six bands of MSG data (0.6, 0.8, 1.6, 3.9, 6.2 and 10.8 渭m) with 10 hidden nodes. The multiple-layer perceptrons lead to a cloud detection accuracy of 88.96%, when trained to map two predefined values that classify cloud and clear sky. The network was further evaluated using sixty MSG images taken at different dates. The network detected not only bright thick clouds but also thin or less bright clouds. The analysis demonstrated the feasibility of using machine learning models of cloud detection in MSG SEVIRI imagery.
    Zhang G. J., 2002: Convective quasi-equilibrium in midlatitude continental environment and its effect on convective parameterization. J. Geophys. Res.,107, ACL 12-1-ACL 12-16, doi: 10.1029/2001JD001005.10.1029/2001JD00100526bee36d9d609af286f08e421e18cea1http%3A%2F%2Fonlinelibrary.wiley.com%2Fdoi%2F10.1029%2F2001JD001005%2Fpdfhttp://onlinelibrary.wiley.com/doi/10.1029/2001JD001005/pdf[1] The quasi-equilibrium assumption proposed by Arakawa and Schubert assumes that convection is controlled by the large-scale forcing in a statistical sense, in such a way that the stabilization of the atmosphere by convection is in quasi-equilibrium with the destabilization by the large-scale forcing. The assumption was developed largely based on observations in the tropical maritime environment and has not been evaluated in midlatitudes. This study examines the quasi-equilibrium assumption in midlatitude continental convection environment using summertime observations from the Southern Great Plains of the United States. Two complementary approaches are taken for this purpose. The first one compares the net time rate of change of convective available potential energy to that due to the large-scale forcing. The second one examines the contributions to the net change of CAPE from the boundary layer air and the free tropospheric air above. Results from both the approaches indicate that the quasi-equilibrium assumption is not well suited for midlatitude continental convection. It is shown that the net change of CAPE is comparable to and largely comes from that due to thermodynamic changes of the boundary layer air, while the contribution from the free troposphere above the boundary layer is negligible. The analysis also shows that the role of convective inhibition to suppress convection is the most pronounced when the large-scale forcing in the free troposphere is weak. On the basis of these and other observations, a modification to the quasi-equilibrium assumption is proposed. It assumes that convective and large-scale processes in the free troposphere above the boundary layer are in balance, so that contribution from the free troposphere to changes in CAPE is negligible. This assumption is then tested using the single column model of the NCAR CCM3 by modifying the closure in the CCM3 convection scheme. Such a modification significantly improves the single column model simulation. The applicability of this new quasi-equilibrium assumption to tropical convection environment is also discussed.
  • [1] Yu DU, Yian SHEN, Guixing CHEN, 2022: Influence of Coastal Marine Boundary Layer Jets on Rainfall in South China, ADVANCES IN ATMOSPHERIC SCIENCES, 39, 782-801.  doi: 10.1007/s00376-021-1195-7
    [2] WANG Xin, Lü Daren, 2005: Retrieval of Water Vapor Profiles with Radio Occultation Measurements Using an Artificial Neural Network, ADVANCES IN ATMOSPHERIC SCIENCES, 22, 759-764.  doi: 10.1007/BF02918719
    [3] XIAO Cunying, HU Xiong, 2010: Applying Artificial Neural Networks to Modeling the Middle Atmosphere, ADVANCES IN ATMOSPHERIC SCIENCES, 27, 883-890.  doi: 10.1007/s00376-009-9019-1
    [4] ZHANG Ming, ZHAO Yanling, HUANG Hong, LIANG Danqing, 2007: The Generalized Energy Equation and Instability in the Two-layer Barotropic Vortex, ADVANCES IN ATMOSPHERIC SCIENCES, 24, 147-151.  doi: 10.1007/s00376-007-0147-1
    [5] Wang Huijun, 2002: The Instability of the East Asian Summer Monsoom-ENSO Relations, ADVANCES IN ATMOSPHERIC SCIENCES, 19, 1-11.  doi: 10.1007/s00376-002-0029-5
    [6] Dong ZHENG, Yijun ZHANG, Qing MENG, Luwen CHEN, Jianru DAN, 2016: Climatology of Lightning Activity in South China and Its Relationships to Precipitation and Convective Available Potential Energy, ADVANCES IN ATMOSPHERIC SCIENCES, 33, 365-376.  doi: 10.1007/s00376-015-5124-5
    [7] Na LI, Baofeng JIAO, Lingkun RAN, Xinyong SHEN, Yanbin QI, 2021: On the Mechanism of a Terrain-Influenced Snow Burst Event during Midwinter in Northeast China, ADVANCES IN ATMOSPHERIC SCIENCES, 38, 800-816.  doi: 10.1007/s00376-020-0104-9
    [8] Shan LU, Zeyong HU, Haipeng YU, Weiwei FAN, Chunwei FU, Di WU, 2021: Changes of Extreme Precipitation and its Associated Mechanisms in Northwest China, ADVANCES IN ATMOSPHERIC SCIENCES, 38, 1665-1681.  doi: 10.1007/s00376-021-0409-3
    [9] Mu Mu, Xiang Jie, 1998: On the Evolution of Finite-amplitude Disturbance to the Barotropic and Baroclinic Quasigeostrophic Flows, ADVANCES IN ATMOSPHERIC SCIENCES, 15, 113-123.  doi: 10.1007/s00376-998-0023-7
    [10] Xiaodan Wu, Cao Hongxing, Andrew Flitman, Wei Fengying, Feng Guolin, 2001: Forecasting Monsoon Precipitation Using Artificial Neural Networks, ADVANCES IN ATMOSPHERIC SCIENCES, 18, 950-958.  doi: 10.1007/s00376-997-0014-0
    [11] John ABBOT, Jennifer MAROHASY, 2012: Application of Artificial Neural Networks to Rainfall Forecasting in Queensland, Australia, ADVANCES IN ATMOSPHERIC SCIENCES, 29, 717-730.  doi: 10.1007/s00376-012-1259-9
    [12] LI Xiangshu, GUO Xueliang, FU Danhong, 2013: TRMM-retrieved Cloud Structure and Evolution of MCSs over the Northern South China Sea and Impacts of CAPE and Vertical Wind Shear, ADVANCES IN ATMOSPHERIC SCIENCES, 30, 77-88.  doi: 10.1007/s00376-012-2055-2
    [13] LU Weisong, SHAO Haiyan, 2003: Generalized Nonlinear Subcritical Symmetric Instability, ADVANCES IN ATMOSPHERIC SCIENCES, 20, 623-630.  doi: 10.1007/BF02915505
    [14] Yong. L. McHall, 1992: Nonlinear Planetary Wave Instability and Blocking, ADVANCES IN ATMOSPHERIC SCIENCES, 9, 173-190.  doi: 10.1007/BF02657508
    [15] Zeng Qingcun, 1989: Variational Principle of Instability of Atmospheric Motions, ADVANCES IN ATMOSPHERIC SCIENCES, 6, 137-172.  doi: 10.1007/BF02658013
    [16] Gao Shouting, 2000: The Instability of the Vortex Sheet along the Shear Line, ADVANCES IN ATMOSPHERIC SCIENCES, 17, 525-537.  doi: 10.1007/s00376-000-0016-7
    [17] Ji Zhongzhen, Lin Wantao, Yang Xiaozhong, 2001: Problems of Nonlinear Computational Instability in Evolution Equations, ADVANCES IN ATMOSPHERIC SCIENCES, 18, 397-403.  doi: 10.1007/BF02919318
    [18] Shen Xinyong, Ni Yunqi, Ding Yihui, 2002: On Problem of Nonlinear Symmetric Instability in Zonal Shear Flow, ADVANCES IN ATMOSPHERIC SCIENCES, 19, 350-364.  doi: 10.1007/s00376-002-0027-7
    [19] Sun Shuqing, 1985: GRAVITY WAVES ON THE AXIS OF LOW LEVEL JET AND THEIR INSTABILITY, ADVANCES IN ATMOSPHERIC SCIENCES, 2, 112-123.  doi: 10.1007/BF03179743
    [20] Li Hongji, Xu Hong, Wang Ronghua, 1988: A HIGH-RESOLUTION ANALYSIS METHOD OF INSTABILITY ENERGY, ADVANCES IN ATMOSPHERIC SCIENCES, 5, 75-86.  doi: 10.1007/BF02657348

Get Citation+

Export:  

Share Article

Manuscript History

Manuscript received: 27 March 2015
Manuscript revised: 17 August 2015
Manuscript accepted: 20 August 2015
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Application of an Artificial Neural Network for a Direct Estimation of Atmospheric Instability from a Next-Generation Imager

  • 1. Department of Atmospheric Science and Engineering, Ewha Women's University, 52 Ewhayeodae-gil, Seodaemun-gu, Seoul 120-750, Republic of Korea

Abstract: Atmospheric instability information derived from satellites plays an important role in short-term weather forecasting, especially the forecasting of severe convective storms. For the next generation of weather satellites for Korea's multi-purpose geostationary satellite program, a new imaging instrument has been developed. Although this imaging instrument is not designed to perform full sounding missions and its capability is limited, its multi-spectral infrared channels provide information on vertical sounding. To take full advantage of the observation data from the much improved spatiotemporal resolution of the imager, the feasibility of an artificial neural network approach for the derivation of the atmospheric instability is investigated. The multi-layer perceptron model with a feed-forward and back-propagation training algorithm shows quite a sensitive response to the selection of the training dataset and model architecture. Through an extensive performance test with a carefully selected training dataset of 7197 independent profiles, the model architectures are selected to be 12, 5000, and 0.3 for the number of hidden nodes, number of epochs, and learning rate, respectively. The selected model gives a mean absolute error, RMSE, and correlation coefficient of 330 J kg-1, 420 J kg-1, and 0.9, respectively. The feasibility is further demonstrated via application of the model to real observation data from a similar instrument that has comparable observation channels with the planned imager.

1. Introduction
  • The prediction of severe weather phenomena, such as tornado-producing thunderstorms, with a sufficient lead-time, is one of the most important tasks for operational weather forecasting. The objective analysis of the pre-convective state of the atmosphere is quite an important task for that purpose. Several indices representing atmospheric instability, such as "Lifted index", "K-index", "Showalter index", "Total Totals", and CAPE, have long been used (Koenig and de Coning, 2009; Botes et al., 2012). These indices are usually derived using the vertical profiles of temperature (T) and humidity (q) obtained from radiosonde observations or the outputs from a numerical weather prediction (NWP) model. However, as radiosonde data are obtained only twice a day at a limited number of observation stations, their spatial and temporal resolutions are limited. Such limitations are severe for a large portion of the globe, including the vast area of the ocean and even over large areas of land. To supplement these limitations, high resolution atmospheric instability indices derived from satellite observations have long been utilized (Kitzmiller and McGovern, 1989; Menzel et al., 1998; Li et al., 2012).

    Satellite-derived instability indices are normally obtained from the vertical T and q profiles derived from the radiances measured by sounding instruments onboard either geostationary or polar orbiting satellites (Menzel et al., 1998; Botes et al., 2012; Li et al., 2012), or by pseudo-sounders onboard geostationary satellites (EUMETSAT, 2013), using statistical (Seemann et al., 2003), physical iterative (Ma et al., 1999), or variational (Koenig and de Coning, 2009) methods. Each observation platform and derivation method has its own advantages and disadvantages. For example, the data from the hyper-spectral sounding instruments onboard the polar orbiters provide the most accurate vertical profiles, although their spatial and temporal resolutions are limited to about 30 km and only twice a day, respectively. On the other hand, the geostationary platform provides a limited accuracy of vertical profiles but with much better spatial and temporal resolutions. For example, the 12-channel Spinning Enhanced Visible and Infrared Imager (SEVIRI) onboard the Meteosat Second Generation (MSG) satellite provides the global instability index (GII) every 15 minutes with spatial resolutions of 15× 15 pixels (corresponding to about 50 km × 50 km) for operation, or 3× 3 pixels for the rapid scan service (Koenig and de Coning, 2009). In terms of retrieval, the physical iteration or variational approach is known to outperform the statistical approach, although its processing time is much longer.

    Here, we introduce a new approach to obtain an instability index, which is developed for a planned series of high performance imagers onboard geostationary satellites, expected to be available as early as in 2015. One of the imagers, the Advanced Meteorological Imager (AMI) will be onboard GK-2A (Geostationary Korea Multipurpose Satellite 2A), the follow-on mission of the current COMS (communication, ocean, and meteorological satellite) program (Kim and Ahn, 2014), and is scheduled to launch in 2018. With the significant improvement of spectral coverage (from the current five channels to sixteen channels, including ten infrared channels), a host of new value-added products, including instability indices, are expected to be produced. Furthermore, as the spatial resolution of the infrared channels will be improved twofold (from 4 km to 2 km), along with the temporal resolution (from 30 minutes to about 10 minutes for the full disk coverage), the available information will increase dramatically. On the other hand, with these improvements, the data volume and burdens for prompt data processing are expected to be significantly increased.

    The new approach is geared toward dealing with this issue, by applying an artificial neural network (ANN) algorithm to derive the instability index directly from the measured radiance (or the brightness temperature, T b), instead of using the retrieved T and q profiles (Koenig and de Coning, 2009; EUMETSAT, 2013). The current algorithm for geostationary satellites, such as Geostationary Operational Environmental Satellite (GOES) Sounders (Jin et al., 2008), produces products with acceptable accuracy by combining observed radiances with NWP forecasts. Furthermore, it is known that most physical iteration or variational approaches converge to a solution within a few iterations, which increases the feasibility for operational applications. However, these approaches require appreciable computation time, which in turn requires certain compromises, such as reduced spatial or temporal resolution of the original observation data (Menzel et al., 1998; Koenig and de Coning, 2009). On the other hand, a recent study by (Li et al., 2012) asserted that satellite-derived stability index is much more useful if the spatial resolution is improved. Also, the improved temporal resolution on the geostationary platform could be better utilized if the derived information is available as soon as the observation data are available. Thus, our choice of the ANN algorithm is to derive the instability index with minimum time delay, and without compromising spatiotemporal resolution. Furthermore, the ANN algorithm could be utilized for cases when high performance NWP data are not available, and has high potential for application to hyperspectral sounding instruments (Liu et al., 2014). However, it should be mentioned here that the sounding information contained within the measured radiance from the geostationary imager is limited, and thus the feasibility of the derived instability index should be carefully validated before real-time application (Martinez et al., 2007).

    The paper is organized as follows: Section 2 describes the specification of the new imaging instrument, along with the preparation of training datasets for the ANN algorithm, including a radiative transfer model (RTM) used for the derivation of the theoretical radiance data with the inputted atmospheric information (i.e., T and q profiles). Section 3 describes the ANN algorithm with the adopted training process and resultant algorithm performance with the training. In section 4, the application results with the actual observation data and its validation are described. The paper concludes in section 5 with some suggestions for future improvements.

2. Data and methods
  • A feed-forward multi-layer perceptron (MLP) algorithm with the back-propagation training approach (Blackwell and Chen, 2009) is adopted for the current study. To properly train the ANN algorithm, it is important to prepare a training dataset with the proper characteristics, such as accuracy, completeness, and comprehensiveness (Blackwell and Chen, 2009). Indeed, a previous study using the pre-sorted limited number of atmospheric profiles from TIGR (Thermodynamic Initial Guess Retrieval) has demonstrated the importance of the comprehensiveness of the training dataset (Lee et al., 2013). Thus, for the current study, the dataset has been augmented with the additional vertical profiles of T and q obtained from the satellite observation, to strengthen the representativeness.

    The training dataset, consisting of input variables such as the measured radiance and viewing conditions, and the corresponding output variable (here, the CAPE value), is prepared by a theoretical approach. For example, the channel radiances are obtained by the convolution of the spectral radiances, [L IASI; see Eq. (1)], obtained from the RTM simulation with the appropriate T and q profiles from the IASI (Infrared Atmospheric Sounding Interferometer) Level 2 (L2) data, with the appropriate boundary and viewing conditions. Here, we use the spectral response function [SRF, φ GEO (Ν)] of the geostationary (GEO) imager, i.e., SEVIRI of MSG, for the latter application. \begin{equation} L_{ GEO/IASI}=\dfrac{\int_\nu{L_{ IASI}(\nu)\phi_{ GEO}(\nu)d\nu}}{\int_\nu{\phi_{ GEO}(\nu)d\nu}} . (1)\end{equation} Here, we use the simulated L IASI instead of the measured IASI Level 1B radiance to simulate the geostationary conditions, including various viewing angles, cloud-free radiances, and boundary conditions that the GEO imager would encounter.

    Once the ANN algorithm is prepared, the optimal way to demonstrate the feasibility would be through the application of the algorithm to the real data. However, as the high performance imager data are not available over the AMI coverage area, we use data from SEVIRI onboard the MSG satellite as a surrogate. Also, as CAPE is the integration over a number of vertical layers and would not be very sensitive to the sounding accuracy of a specific level, as other instability indices would, we focus on CAPE with the expectation that it will properly represent the instability information contained in the planned observational channels.

  • The next generation imaging instrument for GK-2A is known to have much-improved overall capabilities in terms of spatial, temporal, and spectral resolutions, compared to the current five-channel imaging spectrometer onboard COMS (Kim and Ahn, 2014). As summarized in Table 1, its spatial resolution at the sub-satellite point is 0.5 km, 1 km, and 2 km, for the high resolution visible, visible, and infrared channels, respectively, which is at least twofold better than the current imager. Furthermore, the time required to cover a full disk will be much shorter than the current capability of about 30 minutes, by at least threefold, requiring less than 10 minutes for the full disk imaging. Overall, the combined capability is estimated to be more than 30 times better (twofold in spatial, threefold in temporal, and fivefold in spectral terms) than the current instrument.

    There are more interesting and important improvements in view of the atmospheric sounding in the increase of the number of infrared channels with the improved radiometric performance. For example, the water vapor channel will be increased from one channel to three channels, which contain the water vapor information for the lower, middle, and upper troposphere. Furthermore, there will be one new channel in the carbon dioxide absorption band of 13.3 μm, to obtain better temperature information in the upper to middle troposphere. With the combination of these absorption channels and the infrared window channels and a priori information from the NWP model, the instrument is expected to have equivalent or similar vertical pseudo-sounding products compared to the current sounding instruments onboard the U.S. GOES satellite (Schmit et al., 2008; Lee et al., 2014b). However, it should be noted here that the legacy sounding products could be generated with reduced spatial and temporal resolution to limit the computation time and to increase the signal-to-noise ratio (SNR) (Schmit et al., 2008)

  • With 8461 channels covering the infrared spectral range between 3.62 and 15.5 μm, the accuracies of T and q profiles derived from the IASI data (1 K for tropospheric temperature and 10% for humidity) are known to be comparable with those of radiosonde data (Hilton et al., 2012). Thus, IASI L2 T and q profiles in clear-sky conditions in our study domain, (35°-75°N, 75°W-75°E), which corresponds to the rapid scan domain of Meteosat-8, are collected for the training dataset.

    The CAPE values, which are the target products from the ANN algorithm, are calculated using the obtained IASI L2 profiles by integrating the potential temperature differences between the rising air parcel and the ambient air from the level of free convection to the equilibrium level. Normally, the CAPE calculation considers a parcel that is lifted from a well-mixed boundary layer, usually in the lowest 50 to 100 mb, not from the surface (Craven et al., 2002). However, as T and q profiles from both TIGR and IASI L2 represent a certain range of vertical layers, implying that the values could be considered as a layer mean value, we calculate CAPE using the T and q values at the ground, instead of taking average values of the lowest atmospheric layers.

    Figure 1.  (a) CAPE distribution for the collected profiles in the study domain. (b) CAPE distribution for the selected 7197 profiles in the study domain (green: 0-1000 J kg$^-1$; yellow: 1000-2000 J kg$^-1$; orange: $>$2000 J kg$^-1$).

    For sufficient representativeness of the different seasons, IASI L2 T and q profiles are firstly collected from January to December 2012, including about 36 to 40 day and night orbits (or six days) each month. From the period, a total of 459 orbits and 42 025 individual profiles are selected, and then CAPE values are calculated from the profiles. We then filter the collected data based on the calculated CAPE values, to acquire as comprehensive a dataset as possible. For this, we divide the collected data into three groups of CAPE values having "0 and 1000", "1000 and 2000" and "above 2000", and this categorization reveals the natural frequency of CAPE values (as shown in Table 2), with very high frequency of low CAPE values and relatively very low frequency of high CAPE values (i.e., values above 2000, which indicates high instability in the atmosphere). To build up a more extensive training dataset that has sufficient representation at the higher end of the CAPE values, we decided to conduct a second round of data collection, adding more data (63 orbits and 21 852 profiles) taken from summer months. Finally, 63 877 profiles are collected and 2399 profiles from each CAPE range, and a total of 7197 profiles are selected as the final dataset for the training of the neural network.

    Figure 1 shows the distribution of CAPE for the whole of the collected (left) and selected (right) profiles within the study domain. The whole dataset shows a relatively high frequency of low CAPE values (<1000 J kg-1) over most of the study domain, while a high instability toward the regions of southern Europe, between 40°N and 50°N, is evident both in all, and selected, profiles. This distribution agrees quite well with the general pattern of CAPE distribution obtained from the European Centre for Medium-Range Weather Forecasts (ECMWF) 40-Year Reanalysis (ERA-40) dataset, which shows high CAPE values over southern European countries under a dominant influence of the Mediterranean (Romero et al., 2007). Furthermore, Fig. 1 (right) includes an equal number of profiles from three different CAPE ranges, representing three different states of the atmosphere, i.e., weak, moderate, and strong instability.

  • To simulate the theoretical T b from the selected 7197 T and q profiles, we use the most recent version of MODTRAN (MODerate resolution atmospheric TRANsmission, version 5.2.2), which covers spectral ranges from 0.2 to 50 000 cm-1, with 0.2 cm-1 resolution (Berk et al., 2011). Radiances are simulated over the spectral range from 600 to 3000 cm-1 (16.67-3.33 μm), with 1 cm-1 resolution, for each selected T and q profile with the GEO viewing geometry. For the application of Eq. (1), the SEVIRI SRF, available at the EUMETSAT (European Organisation for the Exploitation of Meteorological Satellites) website (http://www.eumetsat.int/website/home/Data/Products/Calibration/MSGCalibration/index.html), is used. Finally, the simulated T b is obtained by the inversion of the Planck function with the center wavelength of each broadband channel. For the independent variables used for the ANN training, T b values from the seven infrared channels (6.2, 7.3, 8.7, 9.6, 10.8, 12.0, and 13.4 μm), along with auxiliary parameters such as geographical location (latitude and longitude), time (converted to the circular type), and satellite zenith angle, are stored.

3. ANN algorithm and training
  • The MLP model has been widely applied in the geosciences, including the atmospheric sciences, such as in areas of pattern recognition, prediction, and function approximation (Gardner and Dorling, 1998; Krasnopolsky, 2007; Taravat et al., 2015). The MLP model used for the current study consists of layers, neurons (or nodes), and connection strength between the nodes (called weights hereafter), and is schematically summarized in Fig. 2. For the number of layers, we selected one input, one hidden, and one output layer, by taking into consideration the nature of complexity (Blackwell and Chen, 2009). In the case of neurons, each layer consists of a different number, while it is relatively easy to determine the input layer (depending on the input variables). However, the neurons in the hidden layer play an important role in the ANN's performance and should be determined during the algorithm's development (see section 3.2 for further detail). On the other hand, the output layer has only one neuron, which combines all the weighted outputs from the hidden layer to produce the estimated value.

    The MLP model used for the current study is a feed-forward (the direction that the information moves) and backward propagation (the direction that the estimated error moves) approach. For the forward direction, the MLP model takes the independent input variables into the input layer and multiplies them by the respective weights, which are the matrix (rows of hidden layers and columns of the input variables). Each weighted input variables is summed and then fed into the hidden layer. In the hidden layer, each neuron activates the input signals using a transfer function, a hyperbolic tangent in our case. The outputs of the hidden layer are then multiplied by the respective weights (i.e., the connection strength between the hidden neuron and the output neuron) and linearly summed up in the output layer to produce the estimated value to complete the forward transfer of the information.

    The other direction of the MLP algorithm, i.e., the backward direction, acts to propagate the error signal, which is obtained by direct comparison between the estimated and target value (the "truth"). This error propagation acts to update the weights that are used in the forward propagation using the error signal, and thus it corresponds essentially to the training of the ANN algorithm. The update is based on the steepest negative gradient with a fixed learning rate (Blackwell and Chen, 2009), which can be expressed as \begin{equation} \label{eq1} w_{ij,{ n}}=w_{ij,{ c}}-\gamma\dfrac{\partial E}{\partial w_{ij}} . (2)\end{equation} Equation (2) states that the current weight (wij, c) is updated by taking into consideration the negative gradient of error (E) relative to each weight (wij) multiplied by a constant, γ, which determines the rate of adjustment (or the learning rate). The new adjusted weights (wij, n) are highly dependent on the sensitivity of the error to each weight and the learning rate value used. Thus, proper selection of the learning rate plays quite an important role in escaping the local minima and speeding up the training process (Blackwell and Chen, 2009), and the sensitivity is tested during the training process to find an optimal value for the current study.

    Figure 2.  MLP model with three layers [$x_i$: inputs; $w_ij$: weight (connection strength) between input neuron $i$ and hidden neuron $j$; $v_j$: weight between hidden neuron $j$ and the output neuron; $f$: transfer function for the hidden neuron; \emphy: the estimation].

  • As shown above, any ANN model requires proper selection of the network architecture, such as the number of layers, neurons, learning rate, and the epoch number. Thus, selection of an optimal combination of these network architecture components is the most important and time consuming process in the training of an ANN model. Following the general procedure suggested by (Lee et al., 2014a), we conducted an extensive performance test for the different combinations of network architectures, including the number of neurons, learning rate, and epoch number, to find an optimal combination. With different numbers of nodes (varying between 9 and 15), different learning rates (increasing from 0.05 to 0.95 with a step of 0.05), and different epoch numbers (from 1000 to 5000 with a step of 1000), a total of 665 combinations of network architectures have been tested. For the test, the training dataset is separated into three groups: a dataset for training (80%), testing (10%), and validation (10%). To prevent any over-fitting problems, we applied the early stopping strategy process (Gardner and Dorling, 1998) with the RMSE value estimated with the validation dataset as the performance indicator. Since the ANN algorithm selects random patterns (e.g., 80 out of 100) for each test-run, the RMSE values vary significantly from one case to another, even with the same combination of ANN parameters. Thus, as mentioned above, a number of tests were conducted to select the best result (minimum RMSE) from each combination.

    Table 3 shows the resultant RMSE values from the different combinations of the numbers of neurons and epochs, with the learning rate of 0.3, which gives the best performance among the values tested. Based on the results, it is clear that the performance is quite sensitive to the selection of network architecture. Furthermore, the RMSE dependence on the selected architecture shows a quite erratic characteristic. For example, with the selection of 12 hidden neurons, the RMSE value shows a gradual decrease with increasing epoch number. However, when the number of hidden neurons is 14, the RMSE value increases with increasing epoch number. On the other hand, when the largest number of epochs is selected, the RMSE value shows a strong variability with the different number of neurons. Nevertheless, the RMSE performance indicates that there is broad agreement on the minimum value of around 420 J kg-1. Thus, for the current study, we selected the combination of 12, 5000, and 0.3, for the number of hidden neurons, the epoch number, and the learning rate, respectively. With this combination, the mean absolute error, RMSE, and cross correlation are 330 J kg-1, 419.9 J kg-1, and 0.9, respectively.

4. Results
  • The feasibility of the ANN algorithm for the real-world data is tested by using the actual data obtained during the rapid scan experiment (2.5 minute image acquisition) of SEVIRI conducted on 20 June 2013, when severe convective storms developed over large parts of Europe (Setvák and Müller, 2013). Figure 3 shows the image sequence of the derived CAPE from the ANN algorithm (hereinafter ANN CAPE) overlaid over the background black and white image of the 10.8 μm T b value (white represents cold clouds, while the dark area is the warm and clear area) at every two hours from 1000 UTC to 1600 UTC (the movie clip made of the retrieved ANN CAPE, with a temporal resolution of 5 min, is available as supplementary online material, entitled Video S1). In the morning hours, several places, such as over the eastern coastal area of Italy, the northern coastal area of Greece, and over Albania, Macedonia, and Bulgaria, show CAPE values of more than 2000 J kg-1. Also, there are large CAPE values along the leading edges of the convective cloud over northern Germany.

    Figure 3.  Sequence of derived CAPE from SEVIRI 2.5 min rapid scan data using the ANN algorithm [displayed every 2 h from 1000 to 1600 UTC: panels (a) to (d), respectively]. The high CAPE values at the leading edges of the developing thunderstorm (red circles) show a continuous development and movement of the developing clouds, while the black circles in (c) represent convective clouds developing later in the afternoon from the clear-sky condition in the morning.

    Figure 4.  (a) ECMWF-forecasted CAPE. (b) ANN CAPE. (c) \emphK-index retrieved from MSG SEVIRI (blank lines indicate missing row data, obtained from the EUMETSAT Data Center). (d) CAPE value derived from the radiosonde observation at 1200 UTC.

    Later on, as the time progresses, several convective clouds over the area where the CAPE values are high and the sky conditions are clear during the morning (black circles), begin to pop up at around 1200 UTC (Fig. 3b). These clouds continue to develop into severe convective clouds in the afternoon hours (Fig. 3c) and begin to weaken during the late afternoon (Fig. 3d), showing an evolution typical of a convective cloud system. On the other hand, the CAPE values along the leading edges of the developing clouds (red circles) are high for a narrow band, while the values at the trailing edges are much weaker. As time progresses, the convective clouds develop and move to the areas with high CAPE values, while the clouds are weakened over the trailing edge (progress of cloud development is clearly visible in the movie clip, Video S1). This kind of CAPE distribution along the developing convective clouds is also shown in the atmospheric instability derived from the sounding instrument onboard the U.S. GOES satellite (e.g., Li et al., 2008, Fig. 2; Lee et al., 2014b, Fig. 9), which qualitatively demonstrate the feasibility of the ANN CAPE. Overall, in a qualitative sense, the ANN CAPE seems to be able to explain the general distribution of the actual atmospheric instability.

  • Here, we attempt to validate the ANN algorithm in terms of the qualitative and quantitative aspects, with the specific case introduced above. For the validation, the ANN CAPE is compared with several similar types of instability information, including the forecasted CAPE from the ECMWF model output (called NWP CAPE), the calculated CAPE using a limited number of radiosonde soundings (Sonde CAPE), and with other atmospheric instability indexes, such as K-index, which is operationally produced from MSG/SEVIRI (EUMETSAT, 2013).

    In the case of the NWP CAPE (Fig. 4a), high instability values are well organized and widely distributed over the northern part of France to Germany, where the severe storm is located in the morning hours. With less strength, there is instability over southeastern parts of Europe, where severe thunderstorms developed later in the afternoon, as shown before (see Fig. 3). Although it is not as strong as the other region, there is also a relatively unstable area in the lower-mid part of Italy, where a single super-cell developed in the afternoon. In some pixels, however, the NWP CAPE is overestimated over the areas where no convective clouds developed (e.g., Slovakia, parts of Austria, and Hungary), and also in cloudy pixels, such as in the northern part of France and Belgium (blue dashed circles). In addition, with the limited spatial resolution of the model (0.25°× 0.25°), the fine-scale description of the local-scale instability, as shown in the satellite-derived instability distribution (Fig. 3b), is limited.

    Comparison with the products from the SEVIRI GII gives a direct comparison between the ANN retrieval and the physical iteration approach. While the GII products are provided every 15 minutes for 3× 3 MSG pixels (or a 9 km × 9 km segment size at the nadir), the ANN retrievals are with 1× 1 pixels. Since CAPE is not included in the GII products, K-index was compared with the ANN CAPE instead. Although the different index scale of K-index (Fig. 4c) and CAPE does not allow a direct quantitative comparison, a careful comparison (Figs. 4b and c) reveals a closer similarity in the spatial distributions of both indices, with a few evident differences in view of the absolute magnitude and exact distribution. While MSG K-index captures the overall instability over Germany well, the ANN products, with higher spatial resolution, are capable of capturing the instability information around cloud edges, e.g., over the northwestern part of Poland (red dashed circle in Fig. 4c) and the rapidly evolving features of small convective cells, as also normally shown for developing super-cell thunderstorms (Li et al., 2008; Lee et al., 2014b).

    The CAPE distribution from the limited number of radiosonde station data obtained from the University of Wyoming (Oolman, 2014), is shown in Fig. 4d. Overall, the areas with high instability are well matched with the high CAPE values from both the ECMWF forecast and ANN algorithm. For example, the strong convective storm that developed over central Europe is led by high Sonde CAPE, which is also shown in NWP CAPE and ANN CAPE. However, due to the limited spatial coverages, several unstable areas, such as over Italy and central Europe, are not captured well by the radiosonde observation. Although it is limited, a quantitative comparison is also made with the radiosonde data available during the time period (1130-1200 UTC), and the results are summarized in Table 4. For the comparison, the high resolution ANN CAPE (about 0.03°× 0.03°, 2.5 min interval) is averaged spatially (within the 0.5°× 0.5° grid-box) and temporally (for 30 min). The limitation in the quantitative validation lies in the scarcity of radiosonde data (e.g., only 44 out of 76 stations have CAPE values at 1200 UTC) and the small number of matches between the retrieved CAPE and that from radiosonde. For example, the high ANN CAPE values along the northeastern coast of Italy (Fig. 4b, red dashed circle) are not validated due to the vacancy of observation (see Fig. 4d). Furthermore, the scarcity of radiosonde stations around the area of convective development, particularly the Mediterranean, including Greece, Bulgaria, and Serbia, limits a comprehensive comparison.

    Even with these limitations, Table 4, which compiles the cases when a sufficient number of retrieved ANN CAPE pixels are available within each comparison grid-box, shows some characteristics of the ANN CAPE. For example, the area- and time-averaged ANN CAPE values are generally much smaller than the collocated Sonde CAPE values, except for a few cases, such as station number 16 080. There are a number of factors that could be considered as the cause of the difference between the two CAPE values. One of the most plausible causes of the difference lies in the limited vertical sounding capability of SEVIRI, especially for the lower atmosphere, both in the amount of information content and its vertical resolution (Jin and Li, 2010). As the accuracy of CAPE estimation is highly sensitive to the lower boundary of the atmosphere (Craven et al., 2002; Zhang, 2002), the lack of information content of SEVIRI for this atmospheric layer could introduce considerable uncertainty. Another possible source for the difference could be introduced by the sampling problem, i.e., the comparison process itself includes several uncertainties, such as the spatial co-location process [radiosondes drift 5 km in the midtroposphere, 20 km in the upper troposphere, and 50 km in the lower stratosphere (Seidel et al., 2011)], and time difference (within one hour). Finally, the satellite signals vertically smoothed by the averaging kernels could reduce the large variability in the vertical T and q distribution, consequently reducing the estimated CAPE value.

    Figure 5.  Time series of the derived ANN CAPE values for a limited area [$2^\circ\times 2^\circ$ box centered at radiosonde station 16080 (black diamond)] at a time interval of $\sim$5 min centered at 1200 UTC.

    Figure 5.  (Continued.)

    Another characteristic of the ANN CAPE is that it shows a large variability even within the small grid-box, having standard deviation ranging from 250 to 760 (Table 4), which might represent the actual spatial variability of the atmospheric instability. To check the spatiotemporal variability of the ANN CAPE, a time series of the high resolution ANN CAPE for a limited area (a 2°× 2° box centered at radiosonde station number 16080) during a one hour time period is shown in Fig. 5 (full temporal resolution images are given in Fig. S1). Within the 5 min time interval, the progress of strong instability along the leading edge of a convective cloud is clearly shown (the convective cloud moves toward the southeast direction, where the high CAPE value is retrieved, as time progresses). Also, there is indeed a great deal of spatial variability in the high resolution ANN CAPE values. For example, although the overall CAPE values within the given area are high (usually at around 1000 J kg-1 in the majority of the area), the highest values, reaching about 4000 J kg-1, are shown in a quite limited band along the cloud edges. Thus, the averaged CAPE value is smaller than the highest value with the larger standard deviation. The spatiotemporal variation of the ANN CAPE over station 14240 shows smaller CAPE values over a large portion of the limited area, with high CAPE values for a limited area, which gives a much smaller ANN CAPE value compared to the Sonde CAPE (see Fig. S2). Nonetheless, the general characteristics (high Sonde CAPE corresponding to relatively high ANN CAPE) demonstrate a qualitative agreement between the two data.

5. Conclusions
  • An ANN algorithm has been developed to derive the atmospheric instability index-specifically, the CAPE-directly from the measured brightness temperature from the planned high performance imager onboard Korea's next-generation geostationary platform. The major benefit of the ANN algorithm over the conventional approach is the production of the atmospheric instability information with high spatial resolution and in a timely manner. With careful preparation of a training dataset, containing as many comprehensive and extensive atmospheric conditions as possible, an MLP model is trained using the back-propagation algorithm. Using the early stopping strategy, a new algorithm consisting of three layers, i.e., input, hidden with 12 neurons, and output layers, with a set of weights giving the least statistical uncertainty, is prepared. The new algorithm is applied to the actual observation data of SEVIRI, a pseudo-sounding imager onboard the geostationary Metosat-8 satellite, to demonstrate its feasibility. Comparisons of the ANN CAPE with other sources of data, such as radiosonde, ECMWF forecast, and MSG-derived instability index, show the possibility of the ANN algorithm to estimate the pre-convective state of the atmosphere. Although the difference in the spatial and temporal resolution between the comparison data limits an extensive and quantitative validation of the algorithm's performance, the overall features of instability over the target area show rather good agreement. For a further feasibility check, we intend to test the effect of the new water vapor channel (6.24 μm for the upper atmospheric layer, 7.34 μm for the middle, and 8.59 μm for the lower), which will be available on the planned AMI, on the performance of the ANN algorithm, with a focus on the East Asian region.

Reference

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return