No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
A satellite infrared technique for estimating “deep/shallow” precipitation
Abstract
This paper describes the statistical characteristics of the Japanese geosynchronous satellite, Geostationary Meteorological Satellite-5 (GMS-5), infrared (IR) data for estimating both “deep/shallow” precipitation. For this study, geographically matched data sets of GMS-5 IR data and composite digital radar data were prepared. By using those data sets, three-dimensional (3-D) matrices of Probability of Rain (PoR) and Mean Rain Rate (mRR) were calculated, where the input variables consisted of three parameters: IR 11 μm brightness temperature (TB11); IR TB difference between 11 μm and 12 μm (TB11–12); and IR TB difference between 11 μm and 6.7 μm (TB11-6.7). The resulting statistical characteristics from those matrices are as follows: •• TB11–12 is an useful parameter for the removal of thin cirrus with no precipitation•• TB11-6.7 is an useful parameter for the extraction of deep convective cloud with heavy precipitationBy using those matrices for looking up PoR and mRR, an empirical algorithm for estimating precipitation was developed. The potential of this technique (denoted as 3-D looking-up table (LUT)) as a now casting tool for severe weather was tested in the case of typhoon “RYAN (T9514)”. The error and scatter of the 3-D LUT estimations were relatively large, but they captured peak rainfalls and accumulative rainfall in good agreement.
... Caso o r calculado seja igual ou superior ao valor do rc, para o grau de liberdade determinado e para o percentual de significância, a hipótese nula é rejeitada e, portanto, se aceita o valor de r como significativo (Bruni, 2007). Kurino (1997). ...
... Portanto, quanto menor for a diferença entre as temperaturas de emissão dos canais 5 e 6, maior será o teor de vapor d'água na camada 500-350 hPa. Da mesma forma, quanto maior for esta diferença, mais seca estará a camada da troposfera.A presença de vapor d'água em níveis elevados está relacionada com a convecção que, ao atingir esses níveis, transporta, verticalmente, o vapor d'água que, por sua vez, aquece o ar sobrepujante, devido ao fato do vapor absorver calor da mesma forma que também libera calor latente através da condensação, como visto nas observações feitas porSchmetzet al. (1997),Kurino (1997) eMecikalski et al. (2010).A Figura 3b representa a distribuição espacial da água precipitável em mm para o mesmo dia e hora, ou seja, 29 de setembro de 2012 às 12h:00min UTC, obtida pela simulação numérica feita com o uso do modelo BRAMS. Pode-se observar que as mesmas áreas mencionadas com baixa diferença de temperatura de emissão pelo vapor d'água correspondem às áreas com alto valor de água precipitável, exceto aquelas áreas sobre a Cordilheira dos Andes, cuja altitude induz baixo valor de água precipitável, visto ser um valor obtido pela integração da umidade específica na vertical. ...
Neste trabalho foi feito uma associação do sensoriamento remoto com a modelagem numérica da atmosfera no propósito de identificar correlações entre imagens do MSG, através dos canais 5 e 6 com a água precipitável, umidade relativa do ar e a diferença entre a temperatura do ar e a temperatura do ponto de orvalho obtidas através do BRAMS. Duas imagens foram estabelecidas para o estudo: 1) Dia 29/09/2012, 12:00 UTC e cortes em 21º S e 27º S; e 2) dia 30/09/2012, 00:00 UTC e cortes em 18º S e 24º S. O modelo BRAMS foi inicializado com dados de re-análises do NCEP/NCAR com saídas a cada 6 horas na mesma resolução espacial e temporal das imagens do MSG (4 km, cada 6 horas). Imagens MSG dos canais 5 e 6 juntamente com o modelo BRAMS se mostraram eficazes no estudo da umidade atmosférica. ABSTRACT This work had the objective of identifying the relationship between MSG satellite data from channels 5 and 6 with precipitable water, relative humidity and the difference between the air temperature and dew point temperature outputs from BRAMS model, in a tentative of associating atmospheric remote sensing and numerical modeling. For this, two base-images were obtained: 1) 29-09-2012, 12:00 UTC ranging between 21°S and 27ºS; and 2) 30-09-2012, 00:00 UTC, ranging between 18°S and 24°S. The model was feeded with NCEP/NCAR Reanalysis data, using the same spatial resolution as the satellite data (4 km in a 6-hour time-step). A 95-99% statistical significance level was obtained throughout the obtained correlations. The MSG satellite data from channels 5 and 6 alongside the BRAMS simulations have proven to be an effective technique in the study of atmospheric humidity. Keywords: BRAMS, Correlations, Images, MSG.
... In both models, one or more of the combination of band WV 6.2 with WV 7.3, IR 8.7, IR 9.7, IR 10.8, IR 12.0, and IR 13.4 (features [16][17][18][19][20][21] was selected in each month. The differences between the brightness temperature in water vapor and IR channels were shown to correspond to deep convective clouds with heavy rainfall [54][55][56]. These band combinations have been proven to provide information about the cloud water path [53] and thus improve rainfall retrieval models [18]. ...
... In both models, one or more of the combination of band WV 6.2 with WV 7.3, IR 8.7, IR 9.7, IR 10.8, IR 12.0, and IR 13.4 (features [16][17][18][19][20][21] was selected in each month. The differences between the brightness temperature in water vapor and IR channels were shown to correspond to deep convective clouds with heavy rainfall [54][55][56]. ...
A new satellite-based technique for rainfall retrieval in high spatio-temporal resolution (3 km, 15 min) for Iran is presented. The algorithm is based on the infrared bands of the Meteosat Second Generation Spinning Enhanced Visible and Infrared Imager (MSG SEVIRI). Random forest models using microwave-only rainfall information of the Integrated Multi-SatEllite Retrieval for the Global Precipitation Measurement (GPM) (IMERG) product as a reference were developed to (i) delineate the rainfall area and (ii) to assign the rainfall rate. The method was validated against independent microwave-only GPM IMERG rainfall data not used for model training. Additionally, the new technique was validated against completely independent gauge station data. The validation results show a promising performance of the new rainfall retrieval technique, especially when compared to the GPM IMERG IR-only rainfall product. The standard verification scored an average Heidke Skill Score of 0.4 for rain area delineation and an average R between 0.1 and 0.7 for rainfall rate assignment, indicating uncertainties for the Lut Desert area and regions with high altitude gradients.
... The increased number of spectral channels of SEVIRI over the previous generation Meteosat sensors makes it possible to develop different multispectral and multithreshold techniques to identify cloud types, such as the split window technique (Inoue, 1987). For example, when the BT (brightness temperature) difference of channels at 11 and 12 µm is greater than 2.5 K, the cloud is considered cirrus (Kurino, 1997). Other authors (Strabala et al., 1994) use BTs in the spectral range of 8-12 µm to identify the cloud thermodynamic phase. ...
... Many studies have focused on identification of storm cells using various satellite data. Kurino (1997) found that the BT difference of 11-6.7 µm is 0 K or less for convective clouds associated with heavy rain. Schmetz et al. (1997) found that the equivalent BT of the 6.7 µm channel can be larger than that of the 11 µm channel by 6-8 K. ...
Identifying deep convection is of paramount importance, as it may be
associated with extreme weather that has significant impact on the
environment, property and the population. A new method, the Hail
Detection Tool (HDT), is described for identifying hail-bearing storms
using multi-spectral Meteosat Second Generation (MSG) data. HDT was
conceived as a two-phase method, in which the first step is the
Convective Mask (CM) algorithm devised for detection of deep
convection, and the second a Hail Detection algorithm (HD) for the
identification of hail-bearing clouds among cumulonimbus systems
detected by CM. Both CM and HD are based on logistic regression models
trained with multi-spectral MSG data-sets comprised of summer
convective events in the middle Ebro Valley between 2006–2010, and
detected by the RGB visualization technique (CM) or C-band weather
radar system of the University of León. By means of the logistic
regression approach, the probability of identifying a cumulonimbus
event with CM or a hail event with HD are computed by exploiting
a proper selection of MSG wavelengths or their combination. A number
of cloud physical properties (liquid water path, optical thickness and
effective cloud drop radius) were used to physically interpret results
of statistical models from a meteorological perspective, using
a method based on these "ingredients." Finally, the HDT was applied
to a new validation sample consisting of events during summer
2011. The overall Probability of Detection (POD) was 76.9% and
False Alarm Ratio 16.7%.
... Later, other spec tral fea tures be gan to be used. For ex am ple, mea sure ments in the range of 6-9 mm are sen si tive to the pres ence of wa ter va por in the at mo sphere [18]. High re flec tivity in the vis i ble range with low tem per a tures in di cates the pres ence of pre cip i ta tion [15]. ...
... Crucial for precipitation estimation is the availability of accurate data, as well as the utilization of suitable techniques to extract significant characteristics from satellite imagery and establish their non-linear correlation with precipitation intensity. In recent years, with the swift progress of computer vision, deep neural networks have offered enhanced and precise techniques for estimating precipitation (Kurino, 1997;Nasrollahi et al., 2013;Stohl & James, 2005). Moreover, they automatically extract the most valuable feature data from extensive data sets without any human involvement (H. ...
Accurate rainfall measurement with a precise spatial and temporal resolution is essential for making informed decisions during disasters and conducting scientific studies, particularly in regions characterized by intricate terrain and limited coverage of automated weather stations. Retrieval of precipitation with satellite is currently the most effective means to obtain precipitation over large‐scale areas. The key to enhancing the accuracy of precipitation estimation and forecasting in regions with complex terrain lies in effectively integrating satellite data with topographic information. This paper introduces a deep learning approach called AttUnet_R_SFT that utilizes high temporal, spatial, and spectral resolution data obtained from the Fengyun 4A satellite, and incorporates the Deep Spatial Feature Transform (SFT) layer to incorporate geographical data for estimating half‐hourly precipitation in northeastern China. We assess it by compared to operational near‐real‐time satellite precipitation products demonstrated to be successful in estimating precipitation and baseline deep learning models. According to the experimental findings, the AttUnet_R_SFT model outperforms practical precipitation products and baseline deep learning models in both identifying and estimating precipitation. The main enhancement of the model performance is shown in the windward slope of the Greater Khingan Mountains as a result of the successful incorporation of geographical data. Hence, the suggested framework holds the capability to function as a superior and dependable satellite‐derived precipitation estimation solution in regions characterized by intricate terrain and infrequent rainfall. The findings of this study indicate that the utilization of deep learning algorithms for satellite precipitation estimation shows potential as a fruitful avenue for further research.
... The time-varying source term can be applied to quantitative precipitation estimation by obtaining ̃ based on physical data. For example, deep convective clouds containing heavy precipitation can be detected using brightness temperature data from geostationary 335 satellites (e.g., Kurino 1997;So & Shin 2018). Generally, larger convective clouds may lead to longer-lasting precipitation. ...
Analytic solutions for the Advection-Diffusion equation have been explored in diverse scientific and engineering domains, aiming to understand transport phenomena, including heat and mass diffusion, along with the movement of water resources. Precipitation, a vital component of water resources, presents a modeling challenge due to the complex interplay between advection-diffusion effects and source terms. This study aims to improve the modeling of nonlinearly evolving precipitation fields by specifically addressing advection-diffusion equations with time-varying source terms. Utilizing analytic solutions derived through the integral transform technique, we modeled the time-varying source term and investigated the correlation between advection-diffusion and source term effects. While the growth of the field is mainly influenced by the amplitude, size, and timescale of the source term, it can be modulated by advection and diffusion effects. When the timescale of source injection is significantly shorter than the dynamic scale of the system, advection and diffusion effects become independent of the field growth. Conversely, when the timescale of source term injection is sufficiently long, the system evolution primarily depends on advection and diffusion effects. In turbulent regimes with strong diffusion and weak advection effects, a quasi-equilibrium state between growth and decay can be established by regulating the decay caused by advection. However, in regimes where advection effects are crucial, the decay process predominates over the growth process.
... Các nghiên cứu này chủ yếu sử dụng dữ liệu của kênh ảnh IR trong dữ liệu vệ tinh theo thời gian thực để truy xuất các sản phẩm mưa dựa trên mối quan hệ lượng mưa với nhiệt độ thu được từ các kênh hồng ngoại. Từ những năm 1997, tác giả [3] đã nghiên cứu sử dụng chênh lệch nhiệt độ đỉnh mây của các kênh IR trong dữ liệu vệ tinh địa tĩnh của Nhật Bản Geostationary Meteorological Satellite-5 (GMS-5) để ước tính lượng mưa. Kết quả nghiên cứu cho thấy sai số và mức độ phân tán lượng mưa tương đối lớn nhưng lượng mưa cực đại và lượng mưa tích lũy ước tính cho kết quả tương đối tốt. ...
... This will cause significant interference with the retrieval model, thus it is vital to minimize cirrus cloud interference as much as possible. On the basis of Kurino [25], who used the bright temperature difference between channels 11 and 12 of the Geostationary Meteorological Satellite (GMS-5) to reduce the interference of upper-level nonprecipitation clouds, the model presented in this paper is trained with the bright temperature difference between channels 13 and 15 of similar wavelength. ...
Meteorological radar data is essential for meteorological monitoring, forecasting, and research, and it plays a crucial role in observing and warning of extreme weather risks. However, meteorological radars have some limitations, such as uneven distribution and severe topographical influence. Meteorological remote sensing satellites can partially overcome these limitations by providing larger observational scope and high spatial and temporal resolution. Using data from meteorological remote sensing satellites to train radar reflectivity factor (RF) retrieval models can effectively compensate for the missing and poor quality of radar data. However, there are still some challenges, such as extracting the features of intense convective weather with unclear coverage from complex multi-channel meteorological remote sensing satellite data and removing the interference caused by non-precipitation clouds on retrieval models. Moreover, the privacy and security of remote sensing data transmission need to be ensured. In this paper, we propose a novel method that combines the Advanced Encryption Standard (AES) method to protect the transmission of remote sensing data, a multi-scale feature fusion module (MFM) to extract multi-scale features from multi-channel meteorological remote sensing satellite data, and an attention technique to reduce the interference of non-precipitation clouds on retrieval models. We conduct comparison experiments with multiple indicators to demonstrate that our method has certain advantages in retrieving radar reflectivity values of different sizes. Our method achieves 0.63, 0.36, 0.49, 0.55 and 0.99 on POD, FAR, CSI, HSS and ACC scores respectively.
... Only IR bands were used, so the AVHRR-based precipitation retrievals can be used during both day and night. The combination of the TB 11 and TB 12 are also known to be effective in detecting thin cirrus clouds that often produce no precipitation (Inoue et al. 1987a,b;Kurino 1997). The second model, hereafter referred to as the AVHRR I, uses TB 11, TB 12, CP, MERRA-2's TWET, and ST as inputs. ...
Precipitation retrieval is a challenging topic, especially in high latitudes (HL), and current precipitation products face ample challenges over these regions. This study investigates the potential of the Advanced Very High-Resolution Radiometer (AVHRR) for snowfall retrieval in HL using CloudSat radar information and machine learning (ML). With all the known limitations, AVHRR observations should be considered for HL snowfall retrieval because (1) AVHRR data have been continuously collected for about four decades on multiple platforms with global coverage, and similar observations will likely continue in the future; (2) current passive microwave satellite precipitation products have several issues over snow and ice surfaces; and (3) good coincident observations between AVHRR and CloudSat are available for training ML algorithms. Using ML, snowfall rate was retrieved from AVHRR’s brightness temperature and cloud probability, as well as auxiliary information provided by numerical reanalysis. The results indicate that the ML-based retrieval algorithm is capable of detection and estimation of snowfall with comparable or better statistical scores than those obtained from the Atmospheric Infrared Sounder (AIRS) and two passive microwave sensors contributing to the Global Precipitation Measurement (GPM) mission constellation. The outcomes also suggest that AVHRR-based snowfall retrievals are spatially and temporally reasonable and can be considered as a quantitatively useful input to the merged precipitation products that require frequent sampling or long-term records.
... The most widely used methods to estimate rainfall from satellite data are based on infrared (IR) sensors, microwave sensors, and precipitation radar ( Kidd and Levizzani, 2011 ). The IR method first discriminates clouds producing rain via a multi-IR channel and then estimates the rain rate ( R ) using the relationship between R and the IR brightness temperature ( T IR ) ( Kurino, 1997 ;Vicente et al., 1998 ;Ba and Gruber, 2001 ). However, this method contains some bias because only cloud-top information is obtained by the IR method. ...
Uncertainties in satellite rainfall estimation may derive from both the local rainfall characteristics and its subpixel variability. To study this issue, Micro Rain Radars and a rain gauge network were deployed within a 9-km satellite pixel in the semi-arid Xilingol grassland of China in summer 2009. The authors characterized the subpixel variability with the coefficient of variation (CV) and evaluated the satellite rainfall estimation for this semi-arid area. The results showed that rainfall events with a high CV were mostly convective with a small amount of rainfall. Spatially inhomogeneous rainfall was most likely to occur at the edges of small clouds producing rain. The performance of the TRMM (Tropical Rainfall Measuring Mission) 3B42 V7 product for daily rainfall was better than that of the CMORPH (Climate Prediction Center morphing technique) and PERSIANN (Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks) products, although the TRMM product tended to overestimate rainfall in a lake area of the semi-arid grassland.
摘要
卫星估雨精度的不确定性受到当地降雨类型和像元内降雨非均匀性影响, 而结合这两个关键因素开展半干旱草原卫星估雨的研究有限.2009年夏, 我们在中国锡林郭勒半干旱草原用多部微雨雷达和雨量计构建了9 km卫星像元降雨观测网, 观测了像元内降雨非均匀性(空间变异系数CV), 并评估了卫星估雨精度.结果表明:(1) CV值受像元内平均降雨量,降雨类型,降雨云面积及移向等影响, 如高CV值的降雨过程大多为平均降雨量小,对流性降雨过程, 降雨云边缘像元CV值较高;(2) TRMM V7卫星估雨产品适用性较好, CMORPH和PERSIANN次之, 但TRMM V7易在半干旱草原湖泊处高估降雨.
... En reprenant une procédure proposée par Kurino (1997) une extension à deux dimensions de l'adaptation d'histogramme est réalisée pour l'ensemble des canaux en infrarouge thermique MSG/SEVIRI. L'adaptation d'histogramme est une méthode simple d'approximation non linéaire couramment utilisé avec une variable explicative. ...
Les sciences de l’atmosphère ont connu une mutation profonde dans les années 1980 avec l’apparition d’un système de satellites météorologiques et la montée en puissance de moyens de traitement. Aussi, les méthodes d’estimation des pluies ont progressivement évolué en relation avec la mise en place de ces nouveaux dispositifs d’observation. Malgré la grande variabilité, dans l’espace et dans le temps, des phénomènes pluviogènes, plusieurs produits opérationnels de mesure des précipitations sont aujourd’hui disponibles. Toutefois ces produits peuvent diverger notablement et leur degré de fiabilité reste difficile à évaluer.Pour analyser la source de ces divergences, ce travail présente synthétiquement ces méthodes de mesure. En premier lieu, des rappels de la physique des précipitations permettront d’évoquer la diversité des divers phénomènes précipitants et de leurs environnements atmosphériques. Puis, à travers quelques exemples d’utilisation, sera mis en évidence la relation entre les échelles spatio-temporelles de résolution et les contraintes opérationnelles associées à un type d’application. Le dispositif de mesure sera ensuite traité dans deux parties. La première sur le dispositif sol traitera aussi bien des mesures directes que de celle fondées sur des propriétés radio-électriques des hydrométéores. La seconde sur les capteurs satellitaires traitera du radar, des micro-ondes passives et de l’infrarouge thermique. Dans ces deux parties les algorithmes d’interprétation des mesures indirectes seront discutés. Enfin, diverses procédures d’élaboration des produits opérationnels seront analysées. La conclusion met en évidence la nécessaire spécialisation des produits de précipitations dont l’efficacité dépend d’un contexte d’exploitation, tenant compte à la fois des caractéristiques physiques du domaine considéré et de la nature de l’information attendue par l’exploitant.
... 174 These segmentalized bands are more useful for monitoring detailed atmospheric 175 characteristics such as humidity, thin ice clouds, and other atmospheric gases (Bessho 176 et al. 2016). The preexisting IRWV methodology, which is useful for diagnosing 177 intense convective clouds (Olander and Velden 2009), was used by Kurino (1997) and 12 Schmetz et al. (1997). The atmospheric window channels generally have warmer 179 brightness temperature than that of the WV channels in the clear-sky and a colder 180 brightness temperature than that in the convective area, which is caused by water 181 vapor reemitting the absorbed radiation from the upper troposphere or lower 182 stratosphere (Schmetz et al. 1997 Tables 3 and 4). ...
This study examines the diurnal variation of the convective area and eye size of thirty rapidly intensifying tropical cyclones (RI TCs) that occurred in the western North Pacific from 2015 to 2017 utilizing Himawari-8 satellite imagery. The convective area can be divided into the active convective area (ACA), mixed-phase, and inactive convective area (IACA) based on specific thresholds of brightness temperature. In general, ACA tends to develop vigorously from late afternoon to early next morning, while mixed-phase and IACA develop during the day. This diurnal pattern indicates the potential for ACA to evolve into mixed-phase or IACA over time. From the thirty samples, RI TCs tend to have at least a single-completed diurnal signal of ACA inside the radius of maximum wind (RMW) during the rapidly intensifying period. In the same period, the RMW also contracts significantly. Meanwhile, more intense storms such as those of category 4 or 5 hurricane intensity are apt to have continuous ACA inside the RMW and maintain eyewall convective clouds. These diurnal patterns of the ACA could vary depending on the impact of large-scale environments such as vertical wind shear, ocean heat content, environmental mesoscale convection, and terrain. The linear regression analysis shows that from the tropical storm stage, RI commences after a slow intensification period, which enhances both the primary circulation and eyewall convective cloud. Finally, after the eye structure appears in satellite imagery, its size changes inversely to the diurnal variation of the convective activity, e.g., the eye size becomes larger during the daytime.
... It is seen that the minimum temperature decreased to 12 UTC, which indicates an increase in the height of the top.The position of the cloud top is found using split technology (split-window technique).Difference between the radiation temperatures of the pixels corresponding to the clouds was determined in the radiometer channels 6.2 and 10.8 μm and then its maximum value was found as max(dif = T6.2 -T10.8); The position of the pixel with the maximum of this difference(Figure 11b) was taken as the position of the top of the cloud[12,13].Almost the same height of cloud top is obtained using 10.8 μm channel itself, where the pixel with minimum radiation temperature is assumed to be associated with cloud top. According toFigure 11f cloud top is increasing from 4 km at 0715 UTC to 17 km at 12 UTC.Preprints (www.preprints.org) ...
The development of extremely powerful thunderstorm which took place on August 19, 2015 is discussed in this paper. High depth hail cloud originated on the Black Sea Coast and classified as a supercell as well as several weaker hailstorms passed more than 1000 km over Northern Caucasus of Russia, the Caspian Sea, and then invaded the territory of Kazakhstan. During more than 20 hours of existence this supercell produced heavy hail, rain, intense lightning discharges, gust and tornado which rarely occurs in the region. The study of the structure and characteristics of the thunderstorm during the formation of electrical discharges and their frequency were of particular interest. According to the forecast, development of convective clouds and separate thunderstorms were expected, though the powerful hail process was not expected due to small vertical temperature gradients and the absence of cold fronts. Supercell was tracked by 5 radars located in this area, which showed its right-hand development with clock-wise deviation from the leading stream on 40-50 degrees to the right and the resulting speed of propagation was about 60-85km/h. The maximum reflectivity factor exceeded value 75dBZ, top of the clouds reached 15-16km and the height of the hail core raised on 11.2km. The size of hailstones size on most of the hail path was 2–3cm, and at the peak of cloud development - 4–5cm. Maximum frequencies of cloud-to-ground flashes of negative and positive polarities reached 30-35min-1 and 60-70min-1 correspondingly, while frequency of cloud-to-cloud flashes was significantly higher and amounted up to 300-500min-1 at the peak of the supercell development. An important fact is that the maximum frequency of flashes of different types coincided in time, showing that the reason of all discharges is similar. Total current of the cloud-to-ground flashes of positive and negative polarities was almost identical in magnitude and differed by sign. It was 200-300 kiloampere at the peak of thunderstorm development. The minimum value of radiation temperature, measured by SEVIRI radiometer installed onboard of Meteosat-10 satellite in 10.8 μm channel, was near to -60ºC. The minimum temperature value on the top of the supercell was comparable to coupled radar and sounding data. The most intensive precipitation flux derived from radiometric measurements was about 22000m3/sec; at the same period radars assessments showed precipitation up to 550mm/h (mixed phase precipitation) and size of hail 4.5cm. The combined satellite-radar-lightning data analysis showed that radar derived characteristics of the supercell reached their maximums earlier than maximum in lightning activity. The highest correlation coefficient between radar and lightning characteristics of the supercell storm was found for pair maximum reflectivity and intensity of LF (0.55) and VHF (0.66) discharges. Estimations of relationship between hail size and lightning activity showed that with increasing hail size, thunderstorm activity increases for both cloud-to-ground and intracloud flashes (on the level 0.46 - 0.59). Analysis of doppler-polarimetric data showed strong inflow zone associated with tornado. Tornadic debris signature was manifested by radar reflectivity factor ZH > 60 dBZ, differential reflectivity ZDR > -1 dB, copolar cross-correlation coefficient ρHV < 0.6, and it was collocated with the tornado vortex signature. Doppler velocities in mesocyclone zone reached values -43 and +63 m/s. Prominent radar echo hook was identified in 1.5 km layer above the ground, while ZDR columns was relatively narrow (4–8 km wide) and not very deep (4.5 km).
... The split window IR channel difference (IR1-IR2) is usually used for discriminating the clouds from aerosol or dust. An IR channel difference (WV-IR1) is useful for recognizing the developing convective clouds (Kurino, 1997); and IR1 channel is useful to detecting the surfaces and clouds. ...
... To date, many efforts have been made to obtain quantitative cloud information that can indicate the evolution of mesoscale convective systems (MCSs). Both Schmetz et al. (1997) and Kurino (1997) proposed methods for estimating convection intensity based on the radiation differences between channels of the Meteosat and GMS-5 geostationary satellites. Machado et al. (1998) studied the morphology and radiative properties of MCSs over the Americas based on year-long GOES-7 satellite observations. ...
In this study, an east-moving Tibetan Plateau vortex (TPV) is analyzed by using the ERA-5 reanalysis and multi-source satellite data, including FengYun-2E, Aqua/MODIS and CALIPSO. The objective is to demonstrate: (i) the usefulness of multi-spectral satellite observations in understanding the evolution of a TPV and the associated rainfall, and (ii) the potential significance of cloud-top quantitative information in improving Southwest China weather forecasts. Results in this study show that the heavy rainfall is caused by the coupling of an east-moving TPV and some low-level weather systems [a Plateau shear line and a Southwest Vortex (SWV)], wherein the TPV is a key component. During the TPV’s life cycle, the rainfall and vortex intensity maintain a significant positive correlation with the convective cloud-top fraction and height within a 2.5◦ radius away from its center. Moreover, its growth is found to be quite sensitive to the cloud phases and particle sizes. In the mature stage when the TPV is coupled with an SWV, an increase of small ice crystal particles and appearance of ring- and U/V-shaped cold cloud-top structures can be seen as the signature of a stronger convection and rainfall enhancement within the TPV. A tropopause folding caused by ageostrophic flows at the upper level may be a key factor in the formation of ring-shaped and U/V-shaped cloud-top structures. Based on these results, we believe that the supplementary quantitative information of an east-moving TPV cloud top collected by multi-spectral satellite observations could help to improve Southwest China short-range/nowcasting weather forecasts.
... Simultaneous observations of these clouds in the infrared window region (11 µm) and the water vapour absorption band (6.7 µm) reveal that the equivalent brightness temperature in the latter is larger than in the former. Negative differences (T B11 -T B6.7 ) are associated with convective clouds with overshooting tops (Kurino 1997). ...
Convective clouds are associated with extreme precipitation events triggering floods. They are an important part of atmospheric circulation and hydrological cycle. Changes in convective clouds in changing climate remain one of the most challenging aspects of forecasting future climate change. The present research focuses on identification of convective clouds using multispectral measurements at split window channels (near 10.5 µm and 12.5 µm) and water vapour absorption channels (near 6.7 µm) from Meteosat 7 observations. Variability of convective clouds has been examined in warming climate using observations from Meteosat First Generation (MFG). It has been reported that convective clouds show high density over Western, Central, North Eastern Indian region, and the Western Ghats during the monsoon period. This observation is consistent with measurement from Precipitation Radar (PR) (reflectivity-based threshold) on-board Tropical Rainfall Measuring Mission (TRMM) and rain gauge-based product. The present technique fails to detect shallow convective clouds over the Western Ghats. An increase of about 32.68% ± 5.81% per degree increase in temperature has been reported in convective clouds over India.
... Figure 11 is similar to Figure 10 but displays the results based on the scenarios in Group 2. MODIS_E (6-15 μm) is better than MODIS (10-12 μm) at rain and snow estimation, indicating that IR channels beyond the commonly used 10-12 μm can also contribute to improvements in precipitation retrieval. The value of water vapor channels (~6.5 μm) has particularly been shown in the past (Ba & Gruber, 2001;Behrangi et al., 2009;Kurino, 1997). These results support joint retrieval using a full range of GOES-16 and GMI channels, although it may not be simple due to differences in observation time. ...
Satellite remote sensing is able to provide information on global rain and snow, but challenges remain in accurate estimation of precipitation rates, particularly in snow retrieval. In this work, the deep neural network (DNN) is applied to estimate rain and snow rates in high latitudes. The reference data for DNN training are provided by two spaceborne radars onboard the GPM Core Observatory and CloudSat. Passive microwave data from the GPM Microwave Imager (GMI), infrared (IR) data from MODIS and environmental data from ECMWF are trained to the spaceborne radar-based reference precipitation. The DNN estimates are compared to data from the Goddard Profiling Algorithm (GPROF) which is used to retrieve passive microwave precipitation for the Global Precipitation Measurement (GPM) mission. First, the DNN-based retrieval method performs well in both training and testing periods. Second, the DNN can reveal the advantages and disadvantages of different channels of GMI and MODIS. Additionally, IR and environmental data can improve precipitation estimation of the DNN, particularly for snowfall. Finally, based on the optimized DNN, rain and snow are estimated in 2017 from orbital GMI brightness temperatures and compared to ERA-Interim and MERRA2 reanalysis data. Evaluation results show that: (1) the DNN can largely mitigate the underestimation of precipitation rates in high latitudes by GPROF; (2) the DNN-based snowfall estimates largely outperform those of GPROF; and (3) the spatial distributions of DNN-based precipitation are closer to reanalysis data. The method and assessment presented in this study could potentially contribute to the substantial improvement of satellite precipitation products in high latitudes.
... These clouds play an important role in the interaction and exchange between troposphere and stratosphere. Simultaneous observations of convective clouds with overshooting tops in the infrared window region (11 µm) and the water vapor absorption band (6.7 µm) show that the equivalent brightness temperature in the latter is larger than in the former (Kurino (1997). Convective clouds with overshooting tops were identified based on negative differences (T B11 -T B6.7 ). ...
... Simultaneous observations of these clouds in the infrared window region (11 µm) and the water vapour absorption band (6.7 µm), reveal that the equivalent brightness temperature in the latter is larger than in the former. Negative differences (TB 11 -TB 6.7 ) are associated with convective clouds with overshooting tops [22]. ...
Convective clouds are the sources of severe weather and extreme precipitation events which often produce flooding, landslides and other disasters. The physical characteristics of convective clouds influence the distribution of radiative heating/cooling in the troposphere. They play a crucial role in atmospheric circulation and the hydrological cycle. Present study deals with the detection of convective clouds using multispectral observations at split window channels (near 11 and 12 µm) and water vapour absorption channels (near 6.7 µm) from EUMETSAT (Meteosat 7) data. Results are compared with the observations (reflectivity-based threshold) from Precipitation Radar (PR) on-board Tropical Rainfall Measuring Mission (TRMM). The Results have also been validated against convective clouds derived from rain gauge based precipitation product from the IMD data. Validation results show a correlation coefficient (cc) of 0.79 and Root Mean Square Error (RMSE) of 2.61 (%) against rain gauge based observations of convective clouds.
... Moreover, water vapour channels, along with channels in the atmospheric window region (8-12 μm), have been commonly used by many authors to discriminate cloud types for water clouds. For example, Kurino (1997a) demonstrated that deep convective clouds tend to have a BTD between 11 and 6.7 μm that is less than or equal to zero. Kurino (1997b) further used three parameters, the brightness temperature at 11 μm, BTD between 11 and 12 μm, and BTD between 11 and 6.7 μm, to calculate the probability of rain and the mean rain rate. ...
Precipitation estimates from satellite infrared (IR) radiometers are typically based on cloud top temperatures. However, these temperatures are weakly related to surface rainfall, particularly for shallow or warm clouds. This study classifies precipitating clouds into five cloud groups. The classification uses three brightness temperature differences (BTDs) and one BTD difference (△BTD) from Himawari‐8 AHI: BTD1 (6.2 ‐ 11.2 μm), BTD2 (8.6 ‐ 11.2 μm), BTD3 (11.2 ‐ 12.4 μm), and △BTD (BTD2 ‐ BTD3). BTD1 is found to be effective for separating shallow and non‐shallow clouds in reference to the Global Precipitation Measurement DPR level 2 data. Once this separation is complete, non‐shallow clouds are further classified. The negative and positive values of △BTD usually indicate more water and more ice in clouds, respectively, distinguishing non‐shallow clouds with tall and taller cloud heights. Subsequently, BTD1 is applied to non‐shallow‐tall/taller clouds. Because these clouds can be considered as optically thick, BTD1 identifies the relative coldness of the cloud top based on the extent of water vapor over the cloud top. The final classification yields four non‐shallow cloud types: non‐shallow‐tall‐cold, non‐shallow‐tall‐colder, non‐shallow‐taller‐cold, and non‐shallow‐taller‐colder clouds. The relationships between IR brightness temperatures and surface rainfall obtained from the classified cloud groups over four latitude bands reveal clear differences, implying that separating cloud types and accounting for regional differences are desirable to improve the accuracy of IR‐based precipitation measurements.
... For the BTD 12.0-10.8 , large negative values are typically used to detect thin clouds such as cirrus, anvils, and so on (Inoue 1985;Kurino 1997;Hong et al. 2010) because the absorption of ice particles in the 12.0-mm channel was stronger than that in the 10.8-mm channel. There are many factors contributing to the changes of BTD 12.0-10.8 ...
Three infrared (IR) indicators were included in this study: The 10.8-μm brightness temperature (BT10.8), the BT difference between 12.0 and 10.8 μm (BTD12.0-10.8), and the BT difference between 6.7 and 10.8 μm (BTD6.7-10.8). Correlations among these IR indicators were investigated using MTSAT-1R images for summer 2007 over East Asia. Temporal, spatial, and numerical frequency distributions were used to represent the correlations. The results showed that large BTD12.0-10.8 values can be observed in the growth of cumulus congestus and associated with the boundary of different terrain where convection was more likely to generate and develop. The results also showed that numerical correlation between any two IR indicators could be expressed by two-dimensional histograms (HT2D). Because of differences in the tropopause heights and in the temperature and water vapor fields, the shapes of the HT2Ds varied with latitude and the type of underlying surface. After carefully analyzing the correlations among the IR indicators, a conceptual model of the convection life cycle was constructed according to these HT2Ds. A new cloud convection index (CCI) was defined with the combination of BTD12.0-10.8 and BTD6.7-10.8 on the basis of the conceptual model. The preliminary test results demonstrated that CCI could effectively identify convective clouds. CCI value and its time trend could reflect the growth or decline of convective clouds.
... 25 MSG SEVIRI imaging can also identify the cirrus clouds, characterized by low temperature in the top layers, but resulting in very low intensity or absence of precipitation. 26,27 The identification and localization of cirrus clouds can result in the reduction of false rainfall alarms. ...
In this work we propose a technique for 15-minutes cumulative rainfall mapping, applied over Tuscany, using Italian weather radar networks together with the regional rain gauge network. In order to assess the accuracy of the radar-based rainfall estimates, we have compared them with spatial coincident rain gauge measurements. Precipitation at ground is our target observable: rain gauge measurements of such parameter have a so small error that we consider it negligible (especially if compared from what retrievable from radars). In order to make comparable the observations given from these two types of sensors, we have collected cumulative rainfall over areas a few tens of kilometres wide. The method used to spatialise rain gauges data has been the Ordinary Block Kriging. In this case the comparison results have shown a good correlation between the cumulative rainfall obtained from the rain gauges and those obtained by the radar measurements. Such results are encouraging in the perspective of using the radar observations for near real time cumulative rainfall nowcasting purposes. In addition the joint use of satellite instruments as SEVIRI sensors on board of MSG-3 satellite can add relevant information on the nature, spatial distribution and temporal evolution of cloudiness over the area under study. For this issue we will analyse several MSG-3 channel images, which are related to cloud physical characteristics or ground features in case of clear sky.
... 25 MSG SEVIRI imaging can also identify the cirrus clouds, characterized by low temperature in the top layers, but resulting in very low intensity or absence of precipitation. 26,27 The identification and localization of cirrus clouds can result in the reduction of false rainfall alarms. ...
The real-time measurement of rainfall is a primary information source for many purposes, such as weather forecasting, flood risk assessment, and landslide prediction and prevention. In this perspective, remote sensing techniques to monitor rainfall fields by means of radar measurements are very useful. In this work, a technique is proposed for the estimation of cumulative rainfall fields averaged over a large area, applied on the Tuscany region using the Italian weather radar network. In order to assess the accuracy of radar-based rainfall estimates, they are compared with coincident spatial rain gauge measurements. Observations are compared with average rainfall over areas as large as a few tens of kilometers. An ordinary block kriging method is applied for rain gauge data spatialization. The comparison between the two types of estimates is used for recalibrating the radar measurements. As a main result, this paper proposes a recalibrated relationship for retrieving precipitation from radar data. The accuracy of the estimate increases when considering larger areas: an area of 900 km2 has a standard deviation of less than few millimeters. This is of interest in particular for extending recalibrated radar relationships over areas where rain gauges are not available. Many applications could benefit from it, from nowcasting for civil protection activities, to hydrogeological risk mitigation or agriculture. © 2015 Society of Photo-Optical Instrumentation Engineers (SPIE).
... To use the information about the cwp and the cloud phase for a proper detection of potentially precipitating cloud areas (i.e., a large enough cwp and ice particles in the upper part of the cloud) the rainfall confidence is calculated as a function of the value combinations of the four variables VIS 0.6 , NIR 1.6 , ΔT 8.7-10.8 , and ΔT 10.8-12.1 (e.g., Bellon et al. 1980;Cheng et al. 1993;Kurino 1997;Nauss and Kokhanovsky 2007). The computation of the pixel based rainfall confidence is realized by a comparison of these combinations with ground based radar data from the German Weather Service (DWD 2005) for daytime precipitation events from January to August 2004 (altogether 850 scenes). ...
The detection of rainfall by geostationary (GEO) weather satellites has a long tradition as they provide area-wide information about the distribution of this key parameter of the water cycle in a very high temporal and high spatial resolution (e.g., Adler and Negri 1988). Most retrieval techniques developed so far for GEO systems are based on the relationship between cloud top temperature in the infrared channel and rainfall probability. Such retrievals which are often referred to as IR retrievals are appropriate for the tropics where precipitation is generally linked with deep convective clouds that can be easily identified in the infrared and/or water vapor channels (e.g., Levizzani et al. 2001; Levizzani 2003) but show considerable drawbacks in the mid-latitudes (e.g., Ebert et al. 2007; Frh et al. 2007) where great parts of the precipitation originates from clouds preferably formed by spatially extended frontal lifting processes in extra-tropical cyclones (hereafter denoted as advective/stratiform precipitation). To overcome this drawback, some authors have suggested to use the effective cloud droplet radius (aef) defined as the ratio of the third to the second power of the cloud droplet spectrum (Hansen and Travis 1974) which can be retrieved from multispectral satellite data. They propose to use values of aef of around 14 ?m as a fixed threshold value (THV) for precipitating clouds (e.g., Rosenfeld and Gutman 1994; Lensky and Rosenfeld 1997; Ba and Gruber 2001) but these studies have mainly focused on convective systems and a fixed THV seems to be not applicable for a reliable differentiation between frontal induced raining and non-raining stratiform clouds over large parts of Europe. In this context, Nauss and Kokhanovsky (2006, 2007) recently proposed a new scheme for the discrimination of raining and non-raining cloud areas applicable to mid-latitudes using daytime multispectral satellite data. Similarly, Thies et al. (2008) introduced a new technique for rain area delineation in the mid-latitudes using night-time multispectral satellite data. In the following sections, the conceptual model of this new approach as well as its application to geostationary MSG (Meteosat Second Generation) SEVIRI (Spinning Enhanced Visible and InfraRed Imager) data will be presented. Since the final technique is different for day- and night-time scenes, the two algorithms will be presented separately.
... The information content of the two channels partially corrects erroneous rainfall area delineation (and consequent frequent rainfall overestimate) of simple IR techniques producing better false alarm ratios (FAR). Kurino (1997a,b) has applied a split-window technique to data from the Japanese Geostationary Meteorological Satellite (GMS). He used three parameters: the 11 µm brightness temperature, the difference between 11 and 12 µm, and the difference between 11 and 6.7 µm. ...
Indirect infrared rainfall estimations were first applied to convective clouds mostly in tropical ocean areas. Raingauge and radar calibration procedures were adopted looking for operational applications, though only recently operational calibration criteria start to be available. Infrared algorithms based on geostationary satellites sensors are still the way to follow for today's operational applications. Statistical analyses, moisture correction factors, and new products from weather radar networks are applied as ancillary data and procedures for improving overall scores. Passive microwave instruments on board polar orbiting spacecrafts are being tested for space-borne calibrations of the infrared techniques and the introduction of hybrid infrared/microwave methods. The expected microwave technological advancements over the next decade lead the way to the first passive microwave instrument on a geostationary orbit. In the mean time multichannel techniques represent a readily available answer for better quantitative estimations; their exploitation requires an adequate training of the forecasters in all physical and meteorological aspects of convective rain genesis, evolution and decay. Satellite precipitation radars represent the future, though their calibration poses difficult problems as for much ground-based weather radars. The coupling of active and passive instruments is expected to amend quantitative methodologies, which still suffer of a considerable degree of approximation.
... A wide variety of studies have been performed to detect the relationship between cloud features and rainfall at the pixel scale. For example, Kurino [2] analyzed the relationship between precipitation and the top brightness temperature (TB) of three infrared (IR) channels of the Geostationary Meteorological Satellite (GMS-5) (IR 11 μm TB (TB 11 ), IR TB difference between 11 μm and 12 μm (TB 11−12 ), and IR TB difference between 11 μm and 6.7 μm (TB 11−6.7 )). Lu and Wu [3] analyzed the precipitation characteristics considering cloud top temperature, temperature gradients and the occurrence of overshooting cloud tops. ...
Although cumulonimbus (Cb) clouds are the main source of precipitation in south China, the relationship between Cb cloud characteristics and precipitation remains unclear. Accordingly, the primary objective of this study was to thoroughly analyze the relationship between Cb cloud features and precipitation both at the pixel and cloud patch scale, and then to apply it in precipitation estimation in the Huaihe River Basin using China’s first operational geostationary meteorological satellite, FengYun-2C (FY-2C), and the hourly precipitation data of 286 gauges from 2007. First, 31 Cb parameters (14 parameters of three pixel features and 17 parameters of four cloud patch features) were extracted based on a Cb tracking method using an artificial neural network (ANN) cloud classification as a pre-processing procedure to identify homogeneous Cb patches. Then, the relationship between Cb cloud properties and precipitation was analyzed and applied in a look-up table algorithm to estimate precipitation. The results were as follows: (1) Precipitation increases first and then declines with increasing values for cold cloud and time evolution parameters, and heavy precipitation may occur not only near the convective center, but also on the front of the Cb clouds on the pixel scale. (2) As for the cloud patch scale, precipitation is typically associated with cold cloud and rough cloud surfaces, whereas the coldest and roughest cloud surfaces do not correspond to the strongest rain. Moreover, rainfall has no obvious relationship with the cloud motion features and varies significantly over different life stages. The involvement of mergers and splits of minor Cb patches is crucial for precipitation processes. (3) The correlation coefficients of the estimated rain rate and gauge rain can reach 0.62 in the cross-validation period and 0.51 in the testing period, which indicates the feasibility of the further application of the relationship in precipitation estimation.
The frequent occurrence of severe convective weather has certain adverse effects on the smart agriculture industry. To enhance the prediction of severe convective weather, the inversion model effectively fills radar reflectivity data gaps by leveraging geostationary satellite data, offering more comprehensive and accurate support for meteorological information in smart agriculture systems. Nevertheless, collaborative cross‐regional inversion driven by dispersed radar data faces challenges in efficiency, privacy, and model accuracy. To this end, we employ an U‐shaped residual network with an embedded light hybrid attention mechanism and utilize a federated averaging algorithm for efficient distributed training across multiple devices which could preserve the privacy of data from different locations, thereby improving inversion performance. In addition, to address the unbalanced nature of radar data, a weighted loss function is designed to enhance the model's sensitivity to high radar reflectivity. Experimental results demonstrate that the proposed model exhibits a certain level of improvement in evaluating radar reflectivity inversion performance across different thresholds compared to other models, thus substantiating the superiority of the proposed approach.
Convolutional neural networks (CNNs) are opening new possibilities in the realm of satellite remote sensing. CNNs are especially useful for capturing the information in spatial patterns that is evident to the human eye but has eluded classical pixelwise retrieval algorithms. However, the black-box nature of CNN predictions makes them difficult to interpret, hindering their trustworthiness. This paper explores a new way to simplify CNNs that allows them to be implemented in a fully transparent and interpretable framework. This clarity is accomplished by moving the inner workings of the CNN out into a feature engineering step and replacing the CNN with a regression model. The specific example of the GOES Radar Estimation via Machine Learning to Inform NWP (GREMLIN) is used to demonstrate that such simplifications are possible and to show the benefits of the interpretable approach. GREMLIN translates images of GOES radiances and lightning into images of radar reflectivity, and previous research used explainable artificial intelligence (XAI) approaches to explain some aspects of how GREMLIN makes predictions. However, the Interpretable GREMLIN model shows that XAI missed several strategies, and XAI does not provide guarantees on how the model will respond when confronted with new scenarios. In contrast, the interpretable model establishes well-defined relationships between inputs and outputs, offering a clear mapping of the spatial context utilized by the CNN to make accurate predictions, and providing guarantees on how the model will respond to new inputs. The significance of this work is that it provides a new approach for developing trustworthy artificial intelligence models.
Significance Statement
Convolutional neural networks (CNNs) are very powerful tools for interpreting and processing satellite imagery. However, the black-box nature of their predictions makes them difficult to interpret, compromising their trustworthiness when applied in the context of high-stakes decision-making. This paper develops an interpretable version of a CNN model, showing that it has similar performance as the original CNN. The interpretable model is analyzed to obtain clear relationships between inputs and outputs, which elucidates the nature of spatial context utilized by CNNs to make accurate predictions. The interpretable model has a well-defined response to inputs, providing guarantees for how it will respond to novel inputs. The significance of this work is that it provides an approach to developing trustworthy artificial intelligence models.
This study proposes a deep-learning-based data-to-data (D2D) translation framework to simulate a radar-like retrieval of rainfall rates using the advantages of spatial coverage and temporal resolution of geostationary (GEO) satellite observation. The D2D method comprises normalization and denormalization in pre- and post-processing and an adversarial learning structure for an inter-domain conversion between physical values of data such as albedo and brightness temperature (BT) unlike the image-to-image translation using digital number values in image data. The GEO-KOMPSAT-2A (GK2A) and radar hybrid surface rainfall (HSR) datasets over the Korean Peninsula from September 2019 to September 2021 were used as the source and target domains for training and testing the D2D model. The constructed D2D model for ground radar-like rainfall generation was validated using the ground radar-observed rainfall data and compared to the GK2A rainfall rate (RR), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Cloud Classification System (PERSIANN-CCS), and Integrated Multi-satellite Retrievals for Global Precipitation Measurement (IMERG) rainfall products. The D2D model exhibited excellent performance for various rain types in the study area compared to the GK2A RR, PERSIANN-CCS, and IMERG data. Consequently, the D2D model can provide valuable and accurate radar-like rainfall intensity and distribution data with a high temporal resolution and complementary rainfall information over lands and oceans without radar observation.
The CRR (Convective Rainfall Rate) algorithm was developed to detect intense mesoscale convective cells and to screen the most probable precipitation associated. It estimates rainfall intensity using the three bands of the Meteosat-7 and matrices calibrated with earth-based radars. Calibration matrices were performed following an accurate version of the Rainsat techniques but combining the infrared bands to detect convective clouds. Matrices were developed, up for the North of Europe, over the Baltic countries, with data from the radar of the Baltex Project provided by the SMHI (Swedish Meteorological and Hydrological Institute) and for the South of Europe, over the Iberian Peninsula, with radar data as provided by the INM (Spanish Meteorological Institute). In the present research, the CRR calibration methodology is validated, an analysis of calibration matrices differences in both areas over Europe is detailed and CRR resulting images are verified in a qualitative manner using rainfall radar images as ground true.
Precipitation estimation with high spatial and temporal resolution is very important for monitoring floods and natural disasters. At present, a couple of quantitative precipitation estimation products and research methods can successfully estimate precipitation at one hourly temporal resolution. In this study, a deep learning model based on Convolutional Neural Network (CNN) was proposed to estimate the precipitation intensity based on the hyperspectral satellite FengYun-4/Advanced Geostationary Radiation Imager (FY-4A), and the temporal resolution is reduced to half an hour. Firstly, the importance of different channels and channel differences for precipitation intensity estimation was determined by ablation experiments. Secondly, compared with the existing model Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks–Convolutional Neural Networks (PERSIANN-CNN) and U-Net. The experimental results show that Small Wisely Network (SW-Net) provides more accurate precipitation intensity estimation, compared with PERSIANN-CNN (U-Net) in the same spatial and temporal resolutions. SW-Net outperformed PERSIANN-CNN (U-Net) by 5.9439% (5.6298%) and 6.3600 (5.8400) percentage points in the loss value and Mean Intersection over Union (MIoU), demonstrating the better feature extraction performance of the model. Furthermore, the False Alarm Ratio (FAR) of precipitation estimation with respect to Global Precipitation Measurement (GPM) Integrated Multi-satellite Retrievals for GPM (GPM-IMERG), for SW-Net was lower than that of PERSIANN-CNN (U-Net) by 49.2132% (49.4302%), showing the higher accuracy of proposed model.
Accurate and timely precipitation estimates are critical for monitoring and forecasting natural disasters such as floods. Despite having high-resolution satellite information, precipitation estimation from remotely sensed data still suffers from methodological limitations. State-of-the-art deep learning algorithms, renowned for their skill in learning accurate patterns within large and complex datasets, appear well suited to the task of precipitation estimation, given the ample amount of high-resolution satellite data. In this study, the effectiveness of applying convolutional neural networks (CNNs) together with the infrared (IR) and water vapor (WV) channels from geostationary satellites for estimating precipitation rate is explored. The proposed model performances are evaluated during summer 2012 and 2013 over central CONUS at the spatial resolution of 0.08° and at an hourly time scale. Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN)–Cloud Classification System (CCS), which is an operational satellite-based product, and PERSIANN–Stacked Denoising Autoencoder (PERSIANN-SDAE) are employed as baseline models. Results demonstrate that the proposed model (PERSIANN-CNN) provides more accurate rainfall estimates compared to the baseline models at various temporal and spatial scales. Specifically, PERSIANN-CNN outperforms PERSIANN-CCS (and PERSIANN-SDAE) by 54% (and 23%) in the critical success index (CSI), demonstrating the detection skills of the model. Furthermore, the root-mean-square error (RMSE) of the rainfall estimates with respect to the National Centers for Environmental Prediction (NCEP) Stage IV gauge–radar data, for PERSIANN-CNN was lower than that of PERSIANN-CCS (PERSIANN-SDAE) by 37% (14%), showing the estimation accuracy of the proposed model.
A new method for the delineation of precipitation during daytime using multispectral satellite data is proposed. The approach is not only applicable to the detection of mainly convective precipitation by means of the commonly used relation between infrared cloud top temperature and rainfall probability but enables also the detection of stratiform precipitation (e.g. in connection with mid-latitude frontal systems). The presented scheme is based on the conceptual model that precipitating clouds are characterized by a combination of particles large enough to fall, an adequate vertical extension (both represented by the cloud water path (cwp)), and the existence of ice particles in the upper part of the cloud. The technique considers the VIS0.6 and the NIR1.6 channel to gain information about the cloud water path. Additionally, the channel differences ΔT8.7-10.8 and ΔT10.8-12.1 are considered to supply information about the cloud phase. Rain area delineation is realized by using a minimum threshold of the rainfall confidence. To obtain a statistical transfer function between the rainfall confidence and the channel differences, the value combination of the four variables is compared to ground based radar data. The retrieval is validated against independent radar data not used for deriving the transfer function and shows an encouraging performance as well as clear improvements compared to existing optical retrieval techniques using only IR thresholds for cloud top temperature.
Because clouds play important roles in producing precipitation and in Earth’s radiative balance, they are a key element in studies of weather and climate, water and energy cycles, and hydrologic analysis. Low clouds have an important effect on cooling the Earth, as they reflect sunlight back to space. High, thin clouds have the opposite effect, allowing incoming sunshine to pass through but trapping heat that is trying to escape from earth. Improving our understanding of cloud structures is the main step in global climate studies and precipitation algorithm development.
Deep convective storms have contributed to airplane accidents, making them a threat to aviation safety. The most common method to identify deep convective clouds (DCCs) is using the brightness temperature difference (BTD) between the atmospheric infrared (IR) window band and the water vapor (WV) absorption band. The effectiveness of the BTD method for DCC detection is highly related to the spectral resolution and signal-to-noise ratio (SNR) of the WV band. In order to understand the sensitivity of BTD to spectral resolution and SNR for DCC detection, a BTD to noise ratio method using the difference between the WV and IR window radiances is developed to assess the uncertainty of DCC identification for different instruments. We examined the case of AirAsia Flight QZ8501. The brightness temperatures (Tbs) over DCCs from this case are simulated for BTD sensitivity studies by a fast forward radiative transfer model with an opaque cloud assumption for both broadband imager (e.g., Multifunction Transport Satellite imager, MTSAT-2 imager) and hyperspectral IR sounder (e.g., Atmospheric Infrared Sounder) instruments; we also examined the relationship between the simulated Tb and the cloud top height. Results show that despite the coarser spatial resolution, BTDs measured by a hyperspectral IR sounder are much more sensitive to high cloud tops than broadband BTDs. As demonstrated in this study, a hyperspectral IR sounder can identify DCCs with better accuracy.
The goal of this paper is to evaluate the feasibility of the Geostationary Operational Environmental Satellite (GOES) Precipitation Index (GPI) technique for the mesoscale convective systems (MCSs) prevailing over the northern part of the South China Sea in the Mei-Yu season. The rain rate retrievals using the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager were used as the oceanic validation data. The global rain threshold, 235 K, and the global GPI coefficient, 3 mm/h, of the GPI technique were adjusted using a method combining the microwave and infrared rain observations from the TRMM satellite. Once the TRMM rainfall observations were not available, IR rain observations from the Japanese Geostationary Meteorological Satellite 5 (GMS-5) provided rainfall information off shore using the adjusted GPI formula. During a period with numerous rainfalls (from 1 May to 12 June 1998), a total of 60 TRMM overpasses were computed in the statistics. Nineteen of these overpasses contained active convections. The average IR rain threshold is 216 K for cases with a spatial averaging scale of 1°. This technique cannot provide adequate rainfall information under such spatial and temporal requirements for the overpasses without active convections. The optimal advantage of the adjusted GPI technique is the simplicity of its calculation and that it demonstrates adequate ability for monitoring MCS induced rainfalls.
The need for frequent observations of precipitation is critical to many hydrological applications. The recently developed high resolution satellite-based precipitation algorithms that generate precipitation estimates at sub-daily scale provide a great potential for such purpose. This chapter describes the concept of developing high resolution Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Cloud Classification System (PERSIANN-CCS). Evaluation of PERSIANN-CCS precipitation is demonstrated through the extreme precipitation events from two hurricanes: Ernesto in 2006 and Katrina in 2005. Finally, the global near real-time precipitation data service through the UNESCO G-WADI data server is introduced. The query functions for viewing and accessing the data are included in the chapter.
Convective clouds are among the most dangerous meteorological phenomena for aviation because they are
responsible of the presence of thunderstorms causing heavy rains, hailstorms, lightnings, wind shear, turbulence and icing phenomena. For this reason it is crucial to detect and early forecast them. In the present work different algorithms have been implemented for reaching this aim using Meteosat Second
Generation satellite data and comparing brightness temperatures of two satellite images in different channels. In addition convective clouds have been simplified in order to report on board only
the relevant information about hazardous areas to be avoided during the flight and to reduce the information weight. In order to provide forecasts in the following 24-48 hours, the algorithms developed have been applied to synthetic satellite images produced by the radiative transfer model RTTOV, a model simulating the radiances and brightness temperatures as they could be seen from satellite
Accurate estimation of global precipitation is critical for the study of the earth in a changing climate. It is generally understood that instantaneous retrieval of precipitation using microwave sensors is more accurate in the tropics and mid latitudes, but the retrievals become difficult and uncertain at higher latitude and over frozen land. In the lack of reliable microwave-based precipitation estimates at high latitudes, retrievals from a single infrared band are commonly used as an alternative to fill the missing gaps. The present study shows that multi-spectral infrared, near-surface air temperature, and near-surface humidity data can add useful information to that obtained from a single infrared band and can significantly improve delineating precipitating from non-precipitating scenes, especially at higher latitudes over land. The role of surface air temperature and humidity is found to be more effective at higher latitudes, but multispectral data is effective across all latitudes. The study is performed using 4 years (2007-2010) of collocated multi-spectral data from the Moderate Resolution Imaging Spectroradiometer (MODIS), surface temperature and humidity data from the European Center for Medium Range Weather Forecast (ECMWF) analysis, and reference precipitation data from CloudSat, which can detect even very light precipitation within 80°S-80°N.
The Moderate Resolution Imaging Spectroradiometer (MODIS) instrument aboard the NASA Earth Observing System (EOS) Aqua and Terra platform with 36 spectral bands provides valuable information about cloud microphysical characteristics and therefore precipitation retrievals. Additionally, CloudSat, selected as a NASA Earth Sciences Systems Pathfinder satellite mission, is equipped with a 94-GHz radar that can detect the occurrence of surface rainfall. The CloudSat radar flies in formation with Aqua with only an average of 60 s delay. The availability of surface rain presence based on CloudSat together with the multispectral capabilities of MODIS makes it possible to create a training dataset to distinguish false rain areas based on their radiances in satellite precipitation products [e.g., Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN)]. The brightness temperatures of six MODIS water vapor and infrared channels are used in this study along with surface rain information from CloudSat to train an artificial neural network model for no-rain recognition. The results suggest a significant improvement in detecting nonprecipitating regions and reducing false identification of precipitation. Also, the results of the case studies of precipitation events during the summer and winter of 2007 over the United States show an accuracy of 77% no-rain identification and 93% detection accuracy, respectively.
Analyzed is the development of three cumulonimbus clouds seeded with the crystallizing reagent in Saudi Arabia. Continuous satellite and radar observations of the clouds were carried out during the five-hour period. Obtained are the data on the dynamics of cumulonimbus clouds and their anvils. The vertical development of cumulonimbus clouds and the increase in the radar reflectivity and amount of precipitation were observed after the seeding. The significant increase in the precipitation was observed in all three cases. The results obtained in the present analysis are in agreement with theoretical concepts of the seeding effects on dynamic properties of clouds and precipitation characteristics. They demonstrate a big potential of seeding for increasing the precipitation falling from cumulonimbus clouds.
The main objective of this study is to analyze the relationship between rainfall intensity and the associated cloud properties which are cloud top temperature (CTT) and cloud cover in Thailand based on some selected case studies during years 2006 and 2007. In addition, the classified cloud data were also applied to the investigation of seasonal cloud and rainfall distribution during those speci-fied years. To assist the efficient derivation of cloud top temperature maps, the automatic cloud classification model for the thermal infrared (TIR) images of the MTSAT-1R satellite was developed and applied as main tool for CTT mapping in the study. And to reduce possible confusion between high clouds and rain clouds (cumulonimbus), the high clouds were filtered off first using the split-window technique under the given thresholds. The classified CTT maps include all clouds with CTT less than 10°C and, as a consequence, most warm clouds and cold clouds are depicted on the obtained maps. The analysis of seasonal cloud and rainfall distribution indicates that patterns of their distri-bution in Thailand are the product of the combined effects among several main driving factors. In summer, these are the local convective system, the cold air mass, the monsoon trough, the westerly wind, and the low pressure area from the ocean. In the rainy season, these are the monsoon trough, the southwest monsoon, and the tropical cyclone and low pressure area from the ocean. And in winter, these are the cold air mass, the northeast monsoon (for the south), and local convection. The amount of total daily rainfall has a high correlation with the amount of cloud cover area seen each day, with r 2 > 0.8 in all cases especially heavy rainfall (e.g. > 80 mm) or on the hail days (with r 2 = 0.8915).
Rainfall intensity was estimated using the MTSAT-1R infrared channels and the microwave satellite precipitation data. Brightness temperature of geostationary satellite is matched temporal and spatial to a variety of microwave satellite(SSM/I, SSMIS, AMSU-B, AMSRE, TRMM) precipitation data. Rainfall intensity was calculated by the look -up table using relationships of MTSAT-1R brightness temperature and microwave precipitation. Estimated rainfall is verified using by precipitation of TRMM satellite(TRMM3B42) and ground rainfall as AWS from Jul. 21 2008 to Jul. 25 2008. The results of rainfall estimated TRMM 2A12(TMI) that validated by AWS and TRMM3B42 precipitation are represented highly 0.38 and 0.61 by correlation coefficient, 5.81 mm/hr and 2.44 mm/hr by RMSE, 0.79 and 0.84 by POD and 0.65 and 0.87 by PC, respectively. Overall, estimated rainfall using by microwave satellite calculated 5 mm/hr or more comparing by AWS and 5 mm/hr or more comparing by TRMM3B42 precipitation, respectively. Validation results of correlation coefficient are shown series of TRMM 2A12, AMSRE, SSM/I, AMSU-B and SSMIS.
Estimates of areal- and time-averaged conective precipitation derived from geostationary satellite imagery using a simple thresholding technique are presented. Three-year means of the estimated precipitation for the period December 1981-November 1984 are shown for each of the (3-month) calendar seasons and compared with published descriptions of the long-term seasonal mean rainfall fields. Over the tropical oceans agreement is quite good with no evidence of any systematic errors. Over the Americas, long-term means derived from station observations of rainfall show less extensive areas of heavy rainfall than those derived here, and a slight tendency for lower peak values during the rainy season. The interannual variability during the 3yr period is described and compared with station observations of rainfall. -from Authors
A simple objective cloud type classification method has been developed, based on split-window measurements of the Advanced Very High Resolution Radiometer on board the NOAA 7 satellite. Brightness temperature difference between the split-window data is a good parameter for the detection of cirrus and blackbody clouds. Two-dimensional histograms of brightness temperature of the 11mum channel and the brightness temperature difference between the split-window data over (64 km)2 subareas are constructed. By selecting appropriate thresholds in the two-dimensional histogram, cirrus, dense cirrus, cumulonimbus, and cumulus clouds are classified over the tropical ocean. Cloud type classification maps were generated by this method for the western Pacific Ocean and were compared with the nephanalysis chart constructed at the Japan Meteorological Satellite Center from GMS data collected within 1 hour of the NOAA 7 observations. The comparison shows reasonable agreement. Fractional cloud cover for cirrus over each (64 km)2 subarea is calculated as the ratio of the number of samples which belong to the cirrus cloud type in the two-dimensional histogram to the number of total samples in the subarea. Fractional cloud cover estimations for cumulonimbus and low-level cumulus are also presented.
A diagnostic method to estimate rainfall over large space and time scales by the use of geosynchronous visible or infrared satellite imagery has been derived and tested. Based on the finding that arms of active convection and rainfall in the tropics are brighter or colder on the satellite visible or infrared photographs than inactive regions, ATS-3 and SMS/GOES images were calibrated with gage-adjusted 10 cm radar data over south Florida. The resulting empirical relationships require a time sequence of cloud area, measured from the satellite images at a specified threshold brightness or temperature to calculate rain volume over a given period. Satellite rain estimates were made for two areas in south Florida that differ in size by an order of magnitude (1.3×104km2 vs 1.1×105km2) and verified by a combined system of gages and radar. Contrary to our expectations, the rain estimates for the smaller area agreed better with the raingage-radar groundtruth than the satellite rain estimates for the large...
A new method of estimating both convective and stratiform precipitation from satellite infrared data is described. This technique defines convective cores and assigns rain rate and rain area to these features based on the infrared brightness temperature and the cloud model approach of Adler and Mack (1984). The method was tested for four south Florida cases during the second Florida Area Cumulus Experiment, and the results are presented and compared with three other satellite rain estimation schemes.
Rain Es~ati~ from Gary satellite ~ag~-v~b~ and i&rat& studies, ~~~. Wea, Rev
- C G Grblth
- P G Wl Woodley
- D W G&e+
- J Martin
- D Stout
- N Siidar
GrBlth, C.G., WL Woodley, P.G. G&e+ D.W. Martin, J. Stout and D,N. Siidar, 1978. Rain Es~ati~ from Gary satellite ~ag~-v~b~ and i&rat& studies, ~~~. Wea, Rev., H&56-75
A cloud type cl~sif~ation with NUAA 7 spilt-wow rn~u~rnen~. f. ~~~~~~ R&r A rainfall estimation with GMS-S infmt& alit-wow and water vapour rn~~rne~ts* ~e~~~~ No@
- T Inot
Inot& T, 1987. A cloud type cl~sif~ation with NUAA 7 spilt-wow rn~u~rnen~. f. ~~~~~~ R&r., 92,3991-Kuritto, T, 1996. A rainfall estimation with GMS-S infmt& alit-wow and water vapour rn~~rne~ts* ~e~~~~ No@, 33, ~~~~ Sat&Be Center fin press)
A rainfall estimation with GMS-5 infrared split-window and water vapour measurements
- Kurino