Rice is the primary food crop in Taiwan. Rice inventory is therefore an important mission for the government to evaluate total rice production and provide a reference to subsidize economic loss caused by agricultural disasters. Currently, a photogrammetric surveying program manually interpreting aerial photos, is employed for the inventory work by the government. The procedures require a large amount of processing time and labor cost. In this research, a new approach to recognizing rice field automatically by integrating geographic information of land parcel and multi-spectral satellite images is proposed. Based on the fact that the vegetation coverage is changing during a rice season, the variation of the spectral properties and the resulting vegetation index provides the information to recognize rice fields. According to this idea, the methods proposed are Differenced-Image­Classification (DIC), Temporal-Profile-Matching (TPM), and Peak-Detecting (PD). In the case of using a single image epoch in paddy classification, the use of geographic information will largely improve the accuracy. When multi-temporal images are used, it can recognize rice fields easily without the need of training data and can further improve the accuracy.

Coastal zone monitoring is an important task in national development and environmental protection. In this the extraction of shorelines is a fundamental work. This investigation aims to extract shorelines in an automated way for optical images. SPOT multi-spectral imagery is the major data source in this investigation for shoreline extraction. The main components of extraction for shorelines include：(1) obtaining a rough separation between the land and water and (2) refining the rough land-water boundaries by edge detection and edge tracing algorithms to extract the accurate shoreline position. To assess the accuracy of shoreline position extracted automatically from the imagery, two validations are included： (1) the comparison of the extracted shorelines to the ones digitized by manual operation and (2) ground measurements by GPS were treated as references. The shoreline positions determined from the images were found to agree with (1) the manual digitization in 2.48% in terms of area, and (2) the ground measurements to within one to two pixels.

A ground-based water vapor radiometer (WVR) was utilized to investigate the inf1uence of water vapor on the propagation of Global Positioning System (GPS) signals through the atmosphere. The influence is in general characterized as so-called“wet delay" (WD). WD were inferred from brightness temperatures at 23.8 and 31.4GHz observed by WVR installed at Taipei weather station from March 18 to 25, 1998, and the corresponding opacity depths. During the field campaign, the WVR was initially installed due east while it was turned to north periodically. An elevation mirror is used to guide atmospheric emission to the antenna. It is designed to rotate perpendicularly to the WVR itself so that elevation angles can be automatically adjusted to scan the atmosphere every 15° at angles between 15° and 165°. In order to derive retrieval coefficients, radiosonde data collected at Taipei weather station in each March starting from 1988 to 1997 were used. WD from WVR measurements varies from 14.8cm for clear sky to 40.8cm for rainy/cloudy conditions at zenith. WD observations at angles from 15° to 165° indicate that stratified atmosphere becomes an less appropriate assumption with viewing angles departing from zenith. The deviation is increased to as much as 8.3cm at angle of 165° .

The atmospheric turbidity is one very important factor in the air pollution measurements and monitoring with remotely sensed data, especially in visible bands. The scattering effects of atmospheric molecular and aerosols in varying atmospheric turbidity conditions can influent the original spectral information of remotely sensed data strongly. From another point of view, the atmospheric turbidity can be estimated by evaluating the information variation induced by the scattering effects. Tanre, et al., proposed the Structure Function (SF) to estimate the atmospheric optical depth in 1988. Their study result showed that the Aerosol Optical Depth (AOD) can be assessed with Landsat TM data by assuming the landcovers are same in the set of multi-temporal TM images. In this study, the SF method is improved for applying in Taiwan area because of the rather rough terrain and complex landuse properties. Besides, the higher spatial resolution SPOT data and hourly GMS-5 data were used to derive the AOD. The result shows the improvements in this study can get satisfying result, and it reveals we can derive these satellite data for the monitoring of hourly air pollution and air quality variation.