应用气象学报

2015, 26(2)

[FullText](2663) [PDF 631KB](697)

Abstract:

Test and Calibration Methods for X-band Active Phased-array Weather Radar

Liu Liping, Wu Chong, Wang Xudong, Ge Runsheng

2015, 26(2): 129-140. DOI: 10.11898/1001-7313.20150201

[FullText HTML] (1258) [PDF 4973KB](2448)

Abstract:
A mobile X-band phased-array meteorological radar (XPAR) is developed by State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, and Anhui Sun-create Electronics Limited Company. The XPAR scans electronically in elevation while scanning mechanically in azimuth, transmits radar wave with wide beam width (about 20° in vertical direction and 1° in horizontal direction) and receives 14 beams simultaneously. As reflectivity calibration is key technique for the active phased-array radar application in meteorological observation, the testing and calibrating method for the XPAR is investigated according to characteristics of the transmitter/receiver (T/R) the multi-beam work mode. The test and calibration focus on the antennas, T/R, purse compress and the variations of gain and beam width with the angle of the antenna beam in respect to the normal of the array face, in order to reduce the observation bias introduced by different modes. After calibration, the XPAR is used to observe 3-D structures and evolutions of convective precipitation in field experiment at Dingyuan of Anhui Province and Ganzi of Sichuang Province from May to August in 2014. The data of an S-band operational radar (SA) and a C-band polarization radar (CPOL) nearby are used to examine the observation capability of the XPAR. Results show that the antenna gain and its variation with the scanning angle, the beam direction, dynamics ranges of T/R are in conformity with the design. The transmitter and receiving characteristics for 128 T/R are similar. The calibration bias for reflectivity and radial velocity measurement are less than 0.98 dB and 0.1 m·s^-1, respectively. Variations of T/R parameters in observation are watched and corrected by the correcting network. Comparing with the SA and CPOL, the bias of reflectivity in Fine Mode is less than 1 dB, the biases for Guard Mode and Quick Mode are less than 2 dB, and the velocity observed in three modes are accordant very well. The bias of reflectivity and radial velocity by XPAR are reasonable. The horizontal and vertical structures of precipitation observed by 3 radars are similar. And calibration results provide basis for quantitative measurement of the XPAR.

Characteristics of Atmospheric Ammonia at Gucheng, a Rural Site on North China Plain in Summer of 2013

Meng Zhaoyang, Xie Yulin, Jia Shihui, Zhang Rui, Lin Weili, Xu Xiaobin, Yang Wen

2015, 26(2): 141-150. DOI: 10.11898/1001-7313.20150202

[FullText HTML] (1250) [PDF 1346KB](1137)

Abstract:
In-situ measurement of ambient ammonia (NH₃) and water-soluble ions in PM_2.5 is conducted at Gucheng, a rural site, from June to August in 2013. Gucheng is an integrated experiment site on ecological and meteorological observation belonging to Chinese Academy of Meteorological Sciences. This station is influenced by high NH₃ emissions from fertilizer use and animal production in surrounding areas. Ammonia and other trace gases are observed by DLT-100 Ammonia Analyzer and a set of commercial instruments during summer of 2013. Hourly concentrations of the water-soluble inorganic ions in PM_2.5 are also measured with the Ambient Ion Monitor (URG 9000 Series, USA). Concentrations of NH₃ at Gucheng range from 0.9×10^-9 to 862.9×10^-9, with the average of 43.9×10^-9±65.9×10^-9. In summer, high temperatures favor ammonia volatilization from fertilizer applied to the cropland. NH₃ concentrations increase sharply after fertilizer application in July for summer maize. Mean concentrations of SO₂, NO_X and O₃ are 4.3×10^-9±5.5×10^-9, 13.2×10^-9±6.8×10^-9 and 42.4×10^-9±31.5×10^-9 during the sampling period. The ammonia shows a significant diurnal variation during the sampling period. NH₃ concentration maximum occurs at 0900 BT and the minimum at 1900 BT. The sulfate, ammonium and nitrate are dominant ions in PM_2.5 with their average concentration being (20.46±13.62), (19.77±33.24) μg·m^-3 and (11.34±9.14) μg·m^-3, respectively. Ammonium shows significant positive correlations with NH₃ concentration. To understand the relationship between particulate ions and their respective precursors, sulfur oxidation ratio (SOR), nitrogen oxidation ratio (NOR) and ammonia conversion ratio (NHR) are investigated. SOR and NOR represent the oxidation ratio of sulfate and nitrate, and NHR represents the conversion ratio of ammonium. Higher SOR and NHR have important effects on the conversion of SO₂ to SO₄^2- and NH₃ to NH₄⁺. The dependence of inorganic PM_2.5 on NH₃ levels suggest that controlling NH₃ emission from agricultural sources could be an efficient way to reduce secondary inorganic particle pollution on North China Plain.

Credibility of Monthly Temperature Predictability Limit and Its Dependence on Length of Data

Liu Jingpeng, Chen Lijuan, Li Weijing, Zhang Peiqun, Zuo Jinqing

2015, 26(2): 151-159. DOI: 10.11898/1001-7313.20150203

[FullText HTML] (1061) [PDF 3177KB](633)

Abstract:
Under the background of global warming, extreme temperature events in China are frequent in recent years, which cause serious influence on economic development and daily life of people. For the evolution of monthly temperature is influenced by initial forcing and boundary forcing, its variation is complex, and brings great challenge to climate prediction. A quantitative investigation is carried out on the monthly temperature predictability limit based on the nonlinear local Lyapunov exponent and daily temperature from 1960 to 2011 at 518 stations in China. But to get robust nonlinear local Lyapunov exponent, there should be enough observations. How much can the length of data series affect the robustness of monthly temperature predictability limit? And what about the credibility of monthly temperature predictability limit? These two questions need to be further analyzed.Based on the nonlinear local Lyapunov exponent and nonlinear error growth dynamics, quantitative analysis is carried out, and it shows that the robustness of monthly temperature predictability limit depends on the length of data series, especially in Northwest China, Northeast China and Central China. In western Inner Mongolia, south of the Yangtze River and South China, data series need to be more than 30 years long. On average, 45-year data series can ensure the stable monthly temperature predictability limit. The length of data series of 518 meteorological stations chosen in this study is 52 years, i.e., they all fit the basic need to evaluate monthly temperature predictability limit. To verify the credibility of monthly temperature predictability limit, the spatial pattern of monthly temperature predictability limit and two objective monthly temperature prediction results are compared. One method is persistent prediction, and the other is monthly dynamic extended range forecast based on climate models. It shows that the spatial distribution of monthly temperature predictability limit and prediction skill is very consistent. The monthly temperature predictability limit evaluated by observation in January is lower than that in July. Similarly, the prediction skill in January is also lower than that in July. What's more, the spatial pattern of objective climate prediction skill in January (July) is similar to the spatial pattern of monthly temperature predictability limit in the respective month. Thus, the monthly temperature predictability limit estimated by nonlinear local Lyapunov exponent and daily temperature from 1960 to 2011 at 518 stations is scientific and credible. And it provides important reference for improvement of monthly temperature prediction.

The Influence of Boundary Layer East Wind on a North China Rainstorm

Wu Qingmei, Liu Zhuo, Wang Guorong, Zhai Liang, Ding Qinglan

2015, 26(2): 160-172. DOI: 10.11898/1001-7313.20150204

[FullText HTML] (1181) [PDF 5496KB](729)

Abstract:
Using conventional observations, 1°×1° NCEP analysis data, ground-based radiometer data, FY-2E meteorolgical satellite and radar data, the boundary layer east wind and its influence on a North China rainstorm on 4 Jun 2013 is analyzed.The boundary layer east wind is from Northeast China Plain, and it becomes moist when passing the Bohai Sea, resulting in cooling in boundary layer, and the sharpest drop is about 9℃ at 925 hPa. The east wind influencing area is within about 300 km. The east wind and according temperature change are monitored accurately by the ground-based radiometer and profile radar, and the storm is triggered after the temperature decreases for about 5 hours.Main influencing weather systems of the rainstorm are the boundary layer east wind, wind shear at mid-low level, southwest low-level jet at 700 hPa and small-scale low trough at 500 hPa. The cold air caused by the boundary layer east wind meets the warm southwest air on the windward area of the Taihang and Yan Mountains, and the cold front is formed near Beijing area. The front lift and topographic lift effects are obvious and the according upward motion is about-0.8 Pa·s^-1, which strengthens upward motion of the warm and moist air near 700 hPa at the north of Beijing. The east wind leads to cooling cushion and temperature inversion at boundary level, and cooling cushion effect triggers the thunderstorm again to some extent, which is generated above the boundary layer, and the most unstable convective available energy reaches 1517.5 J·kg^-1. The elevated thunderstorm is found first to the east of the Taihang Mountains because of topographic lift effect. The analysis of infrared TBB of FY-2E shows that middle convective systems develop obviously when they move near the cold front of east wind. The thunderstorm occurs again just over the east wind cooling cushion area according to radar reflectivity.The moist is sent to the storm area by east winds from boundary layer and southwest winds at mid-low level. The mid-low level warm moist air leads to the increase of stratification convective instability, and at 850 hPa is 8.2 K and 11.7 K more than that of 500 hPa at 0800 BT and 2000 BT, respectively. There is strong dynamic instability over the storm area because the distinct vertical wind shear is formed by boundary layer east winds and strong southwest winds at middle level.

An Improved Bias Removed Method for Precipitation Prediction and Its Application

Sun Jing, Cheng Guangguang, Zhang Xiaoling

2015, 26(2): 173-184. DOI: 10.11898/1001-7313.20150205

[FullText HTML] (1201) [PDF 7672KB](1540)

Abstract:
On the basis of traditional bias removed (BR) method, grading bias removed (GBR) method is designed by adding the step of correcting according to three precipitation orders, which are more than 0.1 mm, 25 mm and 50 mm, respectively. Then, using observations of precipitation and numerical precipitation prediction of ECMWF from April to August in 2011 and 2012, the real-time precipitation forecast of 1-5 days at summer (June-August) over China in 2012 is corrected by GBR method using two different training periods, i.e., the mixed training phase and 60-day running training phase, and the results of them are called GBR_h and GBR_60, respectively. In order to contain information of heavy precipitation in forecast phase as much as possible, the mixed training period is composed of a 30-day period before the forecast phase and two 15-day periods before and after the same phase one year ago, according to characteristics of summer monsoon rainfall of China.Equitable-threat scores (ETS) of forecast over China at many thresholds of precipitation are examined, in order to compare results of the mixed training and the 60-day running training period using GBR. It reveals that both of two corrected results have higher skill than precipitation prediction of ECMWF, at the threshold of beneath 25 mm, the improving amplitude of them are very close (the improvement of GBR_h and GBR_60 are 19.5% and 19.1%, respectively). However, for those above 25 mm, GBR_h apparently has bigger amplitude which is up to 73.5%, and GBR_60 is only 55.9%. Especially in the situation of correcting the local heavy precipitation prediction, the correcting effect of GBR_h is much better. Furthermore, the correlation coefficient is also calculated, and the result shows that the pattern of precipitation prediction is also modified by GBR_h and GBR_60, and the former also has better performance.By analyzing errors of three orders calculated through two different training periods, it is clear that the key point of successfully improving the initial ECMWF forecasts is to add the step of grading bias removed, and a larger improvement of ETS can be expected if more appropriate mixed training period is chosen. It is assumed that according to the obvious effect of this experiment which are easy to apply in operation, this grading bias-removing method of mixed training period will make a very useful product for real time events and have favorable application prospects.

An Algorithm of Optimal Subset for Bayes Precipitation Probability Prediction Model

Hu Banghui, Liu Shanliang, Xi Yan, Wang Xuezhong, You Daming, Zhang Huijun

2015, 26(2): 185-192. DOI: 10.11898/1001-7313.20150206

[FullText HTML] (1177) [PDF 651KB](821)

Abstract:
Based on numerical prediction products, a model output statistic (MOS) for precipitation forecast of an observatory is set up which contains the model output rainfall as one of predictors. The model can remove the systemic error of numerical prediction on precipitation, so it improves the precipitation prediction skill to certain degree. But for a given amount of predictors, a problem to solve is how to select the optimal subset to improve the prediction skill especially in operational weather forecast. In order to construct a Naïve Bayes precipitation probability prediction model on the precondition of the best performance from optimal subsets, using T511 model products and their 13-hour to 24-hour forecast corresponding observation of precipitation from 2008 to 2010 at three observatories, namely Jiexiu, Yuncheng and Fengning, the classificatory Naïve Bayes models on precipitation probability are developed and valuated. Different from the treatment of classic optimal subsets regression which enumerates the optimal subset one by one under the rule of couple score criterion (CSC), a Naïve Bayes model using genetic algorithm to search the optimal subset from a great many of subsets is presented. Model follows artificial intelligence searching characteristics. The genetic algorithm is established through the construction of gene bit-series from binary encoding method, and the introduction of a fitness function with cause. Considering the elimination of non-existing affair samples for the weather of low probability, two models are built based on genetic algorithm and Naïve Bayes model. The essential difference between two kinds of models is the fitness functions they use: One uses the accuracy of precipitation as fitness function, and it is called genetic algorithm-Naïve Bayes forecasting model type 1, GA-NB1 in brief; the other one uses threat score as fitness function, and is called GA-NB2 accordingly. The models are evaluated by prediction tests with dataset ranging from July to September in 2011. Results indicate that simulated results of optimal subset are much superior to those of ordinary initial subsets. Both GA-NB1 and GA-NB2 can improve T511 model precipitation accuracy by 19% on precipitation occurrence, threat scores are improved by 0.16 and 0.13 on drizzle and moderate precipitation, respectively. The prediction for precipitation occurrence and drizzle is enhanced by the optimal subset model because they effectively reduce the false alarm rate of numerical model, by more than 19 times during the period. The cause for improving moderate rain prediction includes two aspects: A slight increase in the amount of correct forecast and decrease of false alarms.

Cloud Influence on Atmospheric Humidity Profile Retrieval by Ground-based Microwave Radiometer

Che Yunfei, Ma Shuqing, Yang Ling, Xing Fenghua, Li Siteng

2015, 26(2): 193-202. DOI: 10.11898/1001-7313.20150207

[FullText HTML] (1221) [PDF 1632KB](978)

Abstract:
There are a lot of limitations on measurement accuracy, cost and continuity of time in the meteorological sounding operations, which are two or four times a day. In order to obtain continuous atmospheric profile data, many methods are developed, among which the way of measuring atmospheric temperature and humidity profiles by the microwave radiometer is relatively mature. However, the ability of the microwave radiometer with infrared sensors is very limited in measuring the cloud, it can only get the height of cloud, and sometimes it brings large deviations. The deviation result in great uncertainty in distributed cloud microwave absorption, causing errors during the inversion of temperature and humidity profiles, so how to improve the accuracy of inversion on cloud is an urgent problem to solve. A method is implemented using atmospheric profiles from L-band sounding radar and brightness temperature observed with microwave radiometer, and MonoRTM is taken as a forward atmospheric radiative transfer model and the tool of retrieval is BP neural network. The matching cloud information is added and a new model of retrieval is created when retrieving atmospheric humidity profiles. Root mean square error (RMSE) values on each height layer with two kinds of inversion method are obtained and the impact of cloud information on atmospheric humidity profile retrieval is analyzed through comparison.Results show that the average of correlation between inversion humidity profiles is improved from 0.6850 to 0.8050 after adding cloud information. Compared with inversion profiles without cloud information, RMSE values on the vast majority of height layers after adding cloud information are reduced to various degrees, which is particularly obvious at layers with cloud.The study shows that the method of adding cloud information on the process of inversion is feasible. In order to improve the ability to observe the atmospheric profile lines in cloudy days, combined information of cloud distribution and brightness temperature of microwave radiation can be used to retrieve the temperature and humidity, in condition the joint observation of cloud radar and microwave radiation is available.

A 3D Self-consistent Propagation Model of the Lightning Leader

Li Dan, Zhang Yijun, Lü Weitao, Chen Lüwen, Zhang Yang, Zheng Dong

2015, 26(2): 203-210. DOI: 10.11898/1001-7313.20150208

[FullText HTML] (1188) [PDF 1238KB](1237)

Abstract:
The research on simulation of lightning discharge and propagation is one of key issues nowadays in the knowledge of lightning physics. To study propagation characteristics of the lightning leader and the interaction between the lightning leader and structures, a 3D self-consistent propagation model of lightning leader is proposed considering the limitation of the 2D random model that cannot describe physical features of lightning discharge three-dimension well. This model proposed can properly simulate the downward stepped leader in 3D space based on the initial conditions, such as the background electric field and the geometry of ground installations. Also, the upward continuous progression of positive connecting leaders from its inception to the final jump as the downward negative stepped leader approaches the ground can be simulated. Meanwhile, this model can self-consistently calculate physical parameters of the leader while it develops in three dimensional space based on the transient electric field under the influence of the downward stepped leader and the geometry of ground structures, such as tip location every step, leader velocity, current density, charge per unit length and the length every step moved. Compared with the optical and electrical data of the natural and artificially triggered lightning detected in Guangzhou experiment base, as well as research results available in literature, simulation results of the self-consistent propagation model show that the initiation of the positive upward leader mainly depends on the electric field intensity and the geometry of the structure in the simulated domain. As the upward leader initiates and propagates towards the downward stepped leader, its velocity increases from 10⁴ m·s^-1 to 10⁵ m·s^-1 order of magnitude gradually along with time during the first 0-600 μs, and then its velocity obviously increases to 10⁶ m·s^-1 order of magnitude when the upward positive leader connects to the downward stepped leader, during which the average velocity reaches about 5×10⁵ m·s^-1. Furthermore, the current intensity in the channel also increases with the upward positive leader moving forward, and the trend has a good coherence with the variation of the relative brightness of the leader, which is consistent with the optical data and existing research. Through simulation results of lightning flashes striking on high structure, it is also derived that the charge per unit length in the propagation channel of upward positive leader initiated from high structure reaches about 50.0-108.0 μC·m^-1, with an average value of 64.3 μC·m^-1. In addition, the average length of upward positive leader is approximately 417 m. The research of 3D self-consistent propagation model of lightning leader can provide reference for further study of characteristics of lightning flashes striking on structures.

A Numerical Study on Characteristics of Cloud-to-ground Lightning Near Surface Configuration

Tan Yongbo, Zhang Dongdong, Zhou Bowen, Shi Zheng, Chen Zhilu, Chen Chao

2015, 26(2): 211-220. DOI: 10.11898/1001-7313.20150209

[FullText HTML] (1228) [PDF 1119KB](1029)

Abstract:
As a common phenomenon in nature, lightning can influence living environment and production extremely. With continuous field lightning observation tests and model experiments, the understanding of lightning process is making a great progress, especially for cloud-to-ground (CG) lightning progression process. The spatial propagation of lightning shows characteristics of randomness, which make lightning unpredictable and lightning protection difficult.The influence of different lightning spatial configurations on CG lightning process is studied, including the location of stroke points, the length of upward leader, the tip location of downward leader when upward leader trigger, form of lightning attachment process. Based on existing model, a region near to the ground is highlighted and the spatial resolution is improved. A 2-dimension model of CG lightning progress process is developed to simulate different lightning spatial configurations by changing random parameters via using the finite difference method. It shows that the difference of lightning spatial configurations will make the location of stroke points different, and random lightning spatial configurations make the length of upward leader random. The range of length of upward leader is 77 m to 609 m, and it concentrates on 100 m to 200 m. Besides, statistical results show that the length of upward leader triggered from building is longer than that triggered from the ground. It also makes the tip location of downward leader when upward leader trigger distribution regularly. The tip location presents ellipsoidal distribution over the building. Also, different lightning spatial configurations will affect the form of CG lightning attachment process. Simulation results emerge three lightning attachment process forms and all can be verified by field lightning observations. All these outcomes show that lightning spatial configuration plays an important role in affecting CG lightning process.In addition, according to a series of statistical analysis, it shows that the length of downward leader and the length of upward leader near the ground have certain linear correlation. The other factors of lightning have little correlations, such as the length of downward leader near the ground and striking distance, the length of upward leader and lightning horizontal extent.

Simulating Evapotranspiration of Rain-fed Soybean Field Based on P-T Model

Wu Wenxin, Jia Zhijun, Dong Yiping

2015, 26(2): 221-230. DOI: 10.11898/1001-7313.20150210

[FullText HTML] (1187) [PDF 1254KB](1124)

Abstract:
Based on eddy covariance measurements and microclimate observations available from 2005 to 2007, the simulating accuracy of evapotranspiration with P-T model of rain-fed soybean field from May to October in Sanjiang Plain is analyzed. Results indicate that simulated values of evapotranspiration by P-T model with conventional parameter (1.26) are significantly higher than observations before emergence and during the growing season of soybean, and the mean bias error (MBE) are 1.65 mm·d^-1 and 1.22 mm·d^-1. However, simulated values are significantly lower than measurements after harvest, with the MBE of-0.74 mm·d^-1. Modeling efficiency (ME) of P-T model are all negative values, which indicates that the model cannot be used in predicting evapotranspiration of soybean field during different periods. The cause may have much to do with the parameter, which is assumed as constant value of 1.26. According to measurements of evapotranspiration, the parameter is derived and shows obviously increasing trend during the whole observation periods. Average values of parameter before emergence, during the growing season, and after harvest are 0.76, 0.86 and 2.20, respectively. It is obvious that the parameter varies according to the growing phase, and it is necessary to modify the parameter based on the measured evapotranspiration of rain-fed soybean field in Sanjiang Plain.Statistical analysis shows that leaf area index (LAI) is an important factor affecting evapotranspiration of soybean field. During the growing season, the parameter is creased with increasing LAI, following a logarithmic equation and a positive correlation. Vapor pressure deficit (VPD) is the direct driving force of transporting vapor from the surface to the surrounding atmosphere. The relationship between and VPD can be described empirically by a piecewise function: When the VPD is greater than 5.05 hPa, it's a positive power function, but when the VPD is lower than 5.05, it's a negative power function. The parameter is positively related to solar radiation and negatively related to VPD before soybean emergency and is positively related to wind speed after soybean harvest.With parameter modified by using linear or non-linear regression equation, the estimation accuracy of P-T model under different periods are improved markedly. Before soybean emergency, MBE and root mean square error (RMSE) are 0.06 mm·d^-1 and 0.60 mm·d^-1, reduced by 96.4% and 71.4%, respectively. ME is improved from a negative to a positive value (0.57), close to the ideal value of 1. During the growing season, MBE and RMSE are 0.15 mm·d^-1 and 0.92 mm·d^-1, reduced by 87.7% and 38.3%, respectively, and ME from a negative to a positive value (0.28). After soybean harvest, MBE and RMSE are-0.21 mm·d^-1 and 0.41 mm·d^-1, reduced by 71.6% and 52.3%, respectively, ME turns from a negative into a positive value (0.42). It indicates that the modified P-T model can simulate the evapotranspiration of soybean field. In conclusion, P-T model is suitable to simulate the evapotranspiration only when the parameter is modified.

Classification of Whole Sky Infrared Cloud Image Using Compressive Sensing

Han Wenyu, Liu Lei, Gao Taichang, Li Yun, Hu Shuai, Zhang Xiaozhong

2015, 26(2): 231-239. DOI: 10.11898/1001-7313.20150211

[FullText HTML] (1232) [PDF 1152KB](998)

Abstract:
Cloud type, as an important macroeconomic parameter in cloud detection, plays a mean role in weather forecasting, field meteorological service, aerospace and climate researches. Automatic identification of cloud types is not efficiently resolved. Cloud shapes, texture, color, contour, range, process of change and some other features are used for manual cloud classification, but it is hard to find a nice way to extract effective features for automatic identification. Particularly, infrared images provide less resolution and less color information.A new method is proposed to classify cloud images obtained from the whole sky infrared cloud measuring system (WSIRCMS) from compressive sensing (CS). Firstly, a redundant dictionary is constructed with typical cloud samples. In order to reduce the computational complexity and computing time, principal component analysis (PCA) and down-sampling is applied to dimension reduction in building up redundant dictionary. It's found that classification results tend to be stable and suitable when the feature contribution rate is more than 95% in PCA or at 16-time down-sampling. Secondly, the optimal solution of paradigm is solved using gradient projection for sparse reconstruction (GPSR) and orthogonal matching pursuit (OMP) algorithms. Sparse algorithm has a certain influence on classification results. There are some negative sparse solutions in GPSR and OMP algorithms, and through the analysis, when the proportion of negative sparse solution is more than 46%, the classification of residual method is prone to error. Sparse solution may be wrong if the incoherence of different type cannot be guaranteed in establishing redundant dictionary, and the dimension reduction may especially increase the correlation. If the cloud texture, structure feature can be kept in process of dimension reduction and one-dimensional treatment and establishing redundant dictionary is complete, it probably makes better sparse solution. Finally, the residual method and sparse proportion method are used to discriminate cloud types. According to experimental results, it's found that the spare-proportion of wave cloud misclassified as cumuliform is less than cumuliform, and for the wave cloud misclassified as stratus cloud or cirrus, its spare-proportion of stratus and cirrus type is small. By combining two discriminated methods, two greatest sparse proportion types are selected and then the small residual is analyzed. Classification accuracy of wave, cumuliform, cirrus cloud is improved.Using compress sensing theory in cloud classification avoids the feature extraction process, and provides a new way for the automatic identification of infrared cloud images. With this method, the recognition rate of waveform, stratiform, cumuliform, cirrus and clear sky reaches 75%, 91%, 70%, 85% and 93%, respectively, with the average accuracy up to 82.8%.

A Comparative Evaluation on Automatic and Manual Observations of Fog and Haze in Tianjin

Si Peng, Gao Runxiang

2015, 26(2): 240-246. DOI: 10.11898/1001-7313.20150212

[FullText HTML] (1123) [PDF 523KB](1097)

Abstract:
Based on the past records by manual observation at ten meteorological stations during 1951-2014 and parallel observations by automatic and manual observations in February 2014, discrepancies of mist, fog and haze phenomena are evaluated in Tianjin, in order to meet the reform direction of surface meteorological observation business, and to improve the quality and availability of data observed by new automatic system. Results indicate that in February, the mean number of mist days in Tianjin is 10, while numbers for fog or haze days are both 2. The proportion for mist accompanying haze days occupies 7.4% in all the weather phenomena, but for fog and haze days, it is only 0.7%. There is an increasing trend in mist days, but not for fog, and since 1990s, haze days have begun to increase obviously, especially for the year of 2007 during the last 60 years. Comparative analysis in parallel period by manual indicates there are 11 more mist days but 6 less fog and haze days than those by automation. Discrepancies of mist and fog phenomena arise at 0800 BT, manual observation captures more mist but automatic observation records more fog. Discrepancies of haze phenomenon arise at 0800, 1400, 1700 BT and 2000 BT, which are observed by manual way, but none according to automatic observations, while the phenomena of haze or mist and haze are totally the opposite. However, characteristics of phenomena judged by automatic and manual observations are of the same, i.e., mist days are the most, haze days are the second, fog days are the least, and the number of mist accompanying haze days is more than that for fog accompanying haze. Comparing visibility values by manual observation with automation indicates that the average relative error between them is 25.1% in all the samples, 54.5% of automatic observation is less than manual observation, 28.5% out of these the difference exceeding 1 km, manual observation meets with automatic observation only for 6.5% of all cases. For differences of visibility in different order of magnitude, 60%-76% of automatic observation is smaller, when the visibility is less than 15.0 km, especially for 0800 BT and 2000 BT. The median of relative error within the threshold ranges of [1.0 km, 5.0 km) and [5.0 km, 10.0 km), judging the phenomenon of mist and haze, are both about 22%, and larger range of [10.0 km, 15.0 km) and [15.0 km, +∞) are only about 15%. Accordingly, observation error of visibility is an important cause leading to discrepancies of mist, fog and haze phenomena by automatic and manual observations, in the case of relative humidity to meet suitable conditions.

Application of the MARS to Data Management of NWP Productions

Xiao Huadong, Sun Jing, Zhang Xi, Bian Xiaofeng

2015, 26(2): 247-256. DOI: 10.11898/1001-7313.20150213

[FullText HTML] (1261) [PDF 1148KB](1309)

Abstract:
The NWP model runs several times every day, creating a large number of files of hundred megabytes each. In the current national meteorological data management system, the NWP model data are managed by the combination of file system and database system, which improves the operation of relative data compared to the traditional file system or database system alone. But data are separated from their index and description, which makes it difficult to manage the massive data of NWP productions.The MARS (Meteorological Archive and Retrieve System) based on the description of meteorology can be used to resolve the above problems. The client-server architecture of the MARS is introduced briefly, and functional components of the MARS are also explained in detail. A data management solution for the national operational NWP model is designed based on the current version of the MARS. According to the characteristic of table-driven code formats, the method of creating GRIB2 code tables and templates is introduced. The application programming interface (API) between the MARS and the Tivoli Storage Manager (TSM) is studied and implemented, considering the I/O feature of disk and tape, and API achieves direct data flushing and retrieving. Taking GRIB2 data of GRAPES global model production for example, data management is implemented by the MARS. Performance test indicates that the MARS can accelerate retrieving of certain data by over one time, demonstrating great advantages of time efficiency.The MARS integrates the meteorological content orient data management technique and the hierarchical storage management technique. Its adaptability and scalability are also verified through long time application in many meteorological centers, showing wider application prospect in the field of NWP production data management.

Vol.26, NO.2, 2015