Chung-Chih Liu

Estimation of the Upper-Layer Rotation and Maximum Wind Speed of Tropical Cyclones via Satellite Imagery

Abstract The movement of convective rainbands embedded in a tropical cyclone (TC) is usually derived from satellite images via the atmospheric motion vector (AMV) method or through the calculation of a radar’s echo track. In estimating the rotation speed of a TC rainband, however, the land-based radar can only detect approaching tropical cyclones within the vicinity. The AMV method is unable to fully account for the TC eyewall movement, thus making it difficult to estimate the TC intensity. The widely used method in estimating the TC maximum wind speed is the Dvorak technique in which the cloud pattern is extracted from only one image. In this study, the rainband rotation speeds are computed via satellite imagery and further applied in estimating the TC maximum wind speed. In contrast to previous research, this study adopts an innovative method by using two subsequent geostationary satellite images. The TC spin rates observed by weather satellites could often be seen to be positively related to the TC int...

Rainfall Intensity Estimation by Ground-Based Dual-Frequency Microwave Radiometers

Abstract Many investigators have used satellite data to derive rainfall intensity and to compare them with rain gauge data. However, there has always been a problem: what is the optimal time period for the two different types of data? A set of well-controlled data collected by ground-based dual-frequency microwave radiometers at the National Central University (24.9°N, 121.1°E) in Taiwan between January of 1996 and December of 1997 was used to find the answer. The results show that a 1-h interval would be the optimal time period and that hourly data will provide a better accuracy than other options (5, 10, or 30 min or 2 h). Two algorithms, the differential and the brightness temperature, were established to estimate rainfall intensity using ground-based dual-frequency microwave brightness temperature and rain gauge data. The results show that the root-mean-square error and the correlation coefficient are 0.63 mm h−1 and 0.88, respectively, for the differential method, and 0.91 mm h−1 and 0.71 for the bri...

A contrast and comparison of near-sea surface air temperature/humidity from GMS and SSM/I data with an improved algorithm

With data sets gained from Defense Meteorological Satellite Program (DMSP) Special Sensor Microwave/Imager (SSM/I) microwave channels, Geostationary Meteorological Satellite (GMS-5) infrared channels, and ship-measured data, the statistical algorithms to estimate sea surface temperature and near-sea surface air humidity around Taiwan and the South China Sea areas are developed. Then a new, improved method to estimate near-sea surface air temperature based on the algorithm proposed by Konda et al. (1996) is established in this study. The results estimated with SSM/I data show that the root mean square error (RMSE) of SST, near-sea surface air humidity and air temperature over the oceans around Taiwan and the South China Sea are 1.2 K,1.43 g/kg, and 1.6 K, respectively. The results with GMS data are 1.7 K,1.71 g/kg and 1.7 K, respectively. The results also show that the improvements in the algorithm of Konda et al. simplify the computation scheme, improve the accuracy, and match the regional ocean-atmosphere properties in retrieving near-sea surface air temperature. The estimate produced using SSM/I and GMS data also show good consistency between them, both in temporal and spatial variations. Basically, the accuracy of this result implies strong potential for application of satellite data to relative studies and operational work in the ocean-atmosphere interaction.

Using Surface Stations to Improve Sounding Retrievals from Hyperspectral Infrared Instruments

Having an accurate atmospheric thermodynamic state is critical for environmental research, particularly the vertical temperature and moisture profiles within the atmospheric boundary layer. This paper investigates the synergistic use of spaceborne hyperspectral infrared radiance measurement and traditional surface observation to conduct the best estimation of atmospheric temperature and water vapor profiles. Comparing the retrieval results from the original spaceborne observation stand-alone algorithm, atmospheric boundary layer temperature and moisture retrievals appear to be improved through the inclusion of the surface observation in the new developed algorithm. The statistics of retrieval performance by comparing with radiosonde observation suggest that the improvement is not only at the lowest surface level but also within the planetary boundary layer. This implies the benefit of surface observation in the atmospheric sounding retrieval algorithm, and the boundary layer thermodynamic structure could be retrieved optimally from the use of both spaceborne and ground-based observations.

Modified Bowen ratio method in near-sea-surface air temperature estimation by using satellite data

The near-sea-surface air temperature can be derived through Special Sensor Microwave/Imager (SSM/I) microwave and Geostationary Meteorological Satellite (GMS-5) infrared data by using a so-called modified Bowen ratio method (MBRM). A specified value (/spl kappa/) that was defined to be a ratio between bulk coefficients c/sub e/ and c/sub h/ was viewed as a constant and was input into this retrieval model for the derivation of near-sea-surface air temperature. However, there is insufficient evidence to prove the hypothesis why the specified value can be viewed as a constant and used in this model. This study strives in providing an answer. By using the iterative technique, the fact that the optimal /spl kappa/ value oscillated merely within a small range for present experimental area covering the area from 10/spl deg/N to 30/spl deg/N and 105/spl deg/E to 135/spl deg/E is shown clearly. The small oscillation does not strongly affect the value of the derived near-sea-surface air temperature. Therefore, the optimal /spl kappa/ value could be considered as a constant both temporally and spatially. Moreover, the results show that the rmse of the estimated near-sea-surface air temperature from SSM/I and GMS-5 data are 1.46 and 1.69 K, respectively. Therefore, the MBRM may be a good approach to estimate this parameter. However, the optimal /spl kappa/ proposed in this study may be suitable only to present experimental area. Generally speaking, this is due to the fact that the /spl kappa/ values are usually larger further to the north (in higher latitudes) than to the south (in lower latitudes) and also during the cold season than in the warm season.

A satellite-derived typhoon intensity index using a deviation angle technique

An objective index was proposed to determine the intensity of typhoons in this study. This was achieved using an image edge processing technique to examine meaningful discontinuity characteristics and thereby calculate the gradient of brightness temperature in satellite infrared images. By taking the typhoon centre as a reference point, the angle between the position vector and the gradient vector was defined as the deviation angle. Following this definition, the probability density and standard deviation of the deviation angle may be derived. After creating a scale from 1 to 0 (0–1) to, respectively, represent the maximum and minimum values of the probability density (standard deviation), this research proposed a non-dimensional typhoon intensity (TI) index. Analysis results reveal a high accuracy when the TI index was used to objectively measure TI. The bias, average error, root mean square error, and R2 value reached 0.6, 3.5, 4.8 m s–1, and 0.89, respectively. Meanwhile, various evaluation parameters in assessing the forecasting skill were also employed, where a specific ‘yes’ and ‘no’ threshold for each typhoon stage was established. The ratio of the number of correct determination to the number of events for a specific typhoon stage was 0.74 (mild), 0.76 (moderate), and 0.89 (severe), respectively, for 557 infrared images of five validation typhoon cases in 2011. The results demonstrated that the TI index technique had good performance in assessing the TI even during typhoon stage changes.

Satellite-derived normalized difference convection index for typhoon observations

Abstract. This study used the spectral features of the geostationary satellite infrared window channel and the water vapor channel data to calculate a new parameter, normalized difference convection index (NDCI), to help determine the overshooting areas in typhoon cloud systems and the centers and intensity of typhoons. The results showed that the two-dimensional NDCI analysis helped to identify typhoon convective cloud systems and the positions of overshooting areas. In addition, because the NDCI values near a typhoon eye were rather significant, if a typhoon eye was formed, the NDCI cross-section analysis could help to confirm its position. When the center of a typhoon was covered by the high anvils and cirrus layers, it could still be qualitatively found through two-dimensional analysis. As for determining the intensity of typhoons, this study also tried to perform correlation analyses with NDCI and maximum sustained wind speed. The result showed that in the ranges within circles of 200 to 250 km radii with a typhoon eye as the center, the correlation between the area with the NDCI values <0 and the maximum sustained wind speed is high with a coefficient 0.7. Thus, the NDCI value could be a referential index to determine the intensity of a typhoon.