Surface temperature observation with nighttime Landsat-5 TM data was conducted at Unzen Volcano, Japan. A series of nine images shows the thermal anomalies of Unzen Volcano from October 1991 through November 1992. In this paper, a method to estimate the surface temperature of the lava dome complex and the surrounding area, and the surface temperature change at Unzen Volcano are discussed.
Surface pixel-integrate temperatures derived from band 5 are generally higher than ones derived from band 7 in the same pixel. It can be explained with a surface temperature model which consist of high temperature areas and low temperature areas within a pixel. If the pixel-integrate temperatures of both bands 5 and 7 are determined for a particular pixel, the temperature and portion of high temperature areas can be estimated for the pixel.
The patterns of temperature distribution which derived from nighttime Landsat TM data correspond to the lava dome and the pyroclastic flow activity of Unzen Volcano.
From the end of 1991 to the beginning of 1992, the high temperature area derived from nighttime Landsat TM at Unzen Volcano decreased. It probably resulted from the decrease in supply rate of lava.

View full abstract

The present paper proposes a method for geometric correction of AVHRR data. A Keplerian model for NOAA satellite and the Bessel ellipsoidal model for the Earth are used. Due to the size of AVHRR imagery, the bilinear interpolation is used to speed up mapping procedure. The optimal block sizes for using interpolation on satellite and map imagery are determined. Because the orbital information on the TBUS still exists tiny errors, there is still some residual error only using THUS. Thus, GCP is usually used to improve the accuracy of geometric correction. The tendency of each orbital element and satellite reported time (satellite internal clock)for the GCP's residual error is analyzed by curve fitting method as a polynomial expression. An algorithm with the polynomial expressions is used to estimate the variations of each orbital element and satellite reported time from the GCP's residual error. Finally, An experimental result is shown to verify the proposed method, and we will conclude our proposed geometric correction method.

View full abstract

[in Japanese]

View full abstract

Aral Sea is a salt lake without outflow rivers laying on the boundary between Kazakhstan and Uzbekistan in Central Asia. The variation of the lake water area mainly depends upon the balance between inflow and evaporation from the water surface. Rapid decreasing of the water area is observed in recent years. The main cause is said to be due to the irrigation for agricultural land in the Amu Dar'ya and the Syr Dar'ya drainage area.
This paper deals with a monitoring of the recent changes of Aral Sea's water area and vegetation area in the drainage basin using multi-temporal satellite data and the other auxiliary sources. A mosaicked LAND SAT MSS and 7 NOAA AVHRR images projected to a Lambert's conformal conic coordinates were used. Normalized difference vegetation index (NDVI) images were obtained using two band data, the visible and near infrared ones. The changes of water area and vegetation area were detected and measured with a thresholding method. Finally, the authors attempted to grasp the changes of the hydrologic environment in the Aral Sea area by analyzing changes of the water area and vegetation area.
Changes of Aral Sea area and its hydrologic environment were summarized as follows ;
1) By 1992, the water area proved to be less than 55% by comparison with that of the statistic yearbook of 1960's.
2) Shoreline indicated remarkable changes in the east and south lake side. However, the changes could not be clearly seen along the west shoreline because the west side touches the steep cliff of the plateau.
3) A decrease in the water area and an increase in vegetation area along rivers with the lapse of time were recognized. This relation could been matched with the influence of the irrigation for the agricultural land from rivers.
4) The changes in hydrologic environment of Aral Sea were analyzed by using multi-temporal satellite data. A decrease in water volume was observed, and decreases in evaporation and inflow volume were estimated.
5) The decrease in river water toward the river mouth from the upstream was shown clearly by measuring the river width on the MSS images. A lot of water loss due to irrigation in the deltas was proved.

View full abstract

A sea surface effect (SSE), in this paper, is meant as the temperature difference between the uppermost sea surface and a certain depth for sea truth. The SSE is known as one of major error sources in the sea surface temperature (SST) estimation by remote sensing. In order to investigate the SSE, a special measurement device called a SST profiler buoy (SSTPB) was developed and has been installed in Must Bay for two years.
This paper is concerned with the statistical analysis for the observed data by SSTPB. It is interpreted that the SSE is apparent when the temperature difference between the uppermost sea surface and the 1 m depth is more than 0.5^°C. The maximum of the difference was 4.6^°C observed at 14 : 30 on July 7, 1992. The SSE appeared consistently in the calm and clear daytime from May to September. It grows when the solar elevation is higher than 36 deg. and the wind speed is less than 2 m/s. The rate of occurrence was about 20% of the total days of observation.

View full abstract