| Summary: | 博士 === 國立交通大學 === 土木工程系所 === 106 === This study aims to monitor the glacier height changes by T/P-series altimeters (Topex/Posedion (T/P), Jason-1 (J1), Jason-2 (J2) and Jason-3 (J3)) in Alaska and Tibet. Such as Antarctic and Arctic, large parts of Alaska and Tibet are covered by glaciers. The rises of temperatue have accelerated glacier melting and lead to the increase in global sea level. In addition, the Tibet Plateau is known as Asia’s water tower. Glacier losses in Tibet will affect the economy and life in the downstream areas with rivers originating from Tibet. Measuring glacier height changes in Alaska and Tibet by radar altimetry can be difficult. The footprint of the radar illuminated area can exceed 1 km in radius, and the radar waveform can be corrupted by complex glacier surface, and the radar range measurement can be biased by steep terrain. In this study, a glacier-threshold method (GTM) is used to generate reliable height change measurements over mountain glaciers in Alaska and Tibet. The computational procedure of GTM for T/P-series satellite altimeters can be divided into five steps. (1) Retrieving Ku-band ranges, satellite altitudes, returned waveforms, and geophysical corrections, and then improving returned range by using waveform retracking. The best algorithm of waveform Retracking is determined by the result of crossover analysis. (2) Determining whether radar altimetry footprints are over glaciers by using Global Land Ice Measurements from Space (GLIMS) and Google Earth images. (3) Editing and selecting altimetry observations using waveform classification and DEM. (4) Reducing the effects of terrain and correcting terrain gradient errors. The best method of terrain gradient correction is again based on crossover analysis. (5) Removing outliers and smoothing height changes to form the final time series of glacier height change. The sub-waveform retracker (50% threshold value) results in a minimum difference (0.01 m/year) between the height change rates at a crossover point in Alaska. Compared to a surface from fitting height observations, SRTM results in more precise height changes when applying terrain gradient correction. GTM removes questionable heights observations and slects at least 15% of original data for detecting glacier height changes in Alaska and Tibet. In Alaska, this study computes time series of height change from T/P (1993-2002) at 20 sites, and at 47 sites from J2 (2008-2016). The J2 result indicates that most height changes in Alaska glaciers have negative rates (meaning melting). The largest melting rate is 11.06±0.35 m/year at Klutlan Glacier. Some of the rates are positive (meaning thickening). The largest thickening rate is 5.99±0.92 m/year at Fan Glacier. In Tibet, this study detects glacier height changes at Mt. Tanggula and Mt. Kunlun (central and western Tibet), and obtains 8 time series from T/P. The J2 observations over the two istes have a poor data quality, thus producing no result. The rates of height change at Site A and B of Mt. Tanggula glaciers are -3.57±0.15 and -1.36±0.15 m/year, respectively. A time series of lake level change of Chibuzhang Co is determined by the data from the T/P-series altimetry satellites. This study analyzes the correlation between the height change of Mt. Tanggula glaciers and lake level change of Chibuzhang Co. The peak of glacier height in Mt. Tanggula occurred in winter, while the peak of lake level occurred in summer. From T/P, the elevations at the ice tongues of central Mt. Kunlun decreased: the rates at Site C, G and H are -1.52±0.76, -5.72±0.79 and -1.11±0.57 m/year. During 1997-98, anomalous changes in glacier height and lake level occurred at Mt. Tanggula and Chibuzhang Co. These anoamlies may be associated with the 1997-98 EL Niño. GTM can effectively select the best altimeter measurements over rugged glaciers to monitor their height changes. T/P-series radar altimeters are continuous missions spanning 25 years. The mission is J3, launched on January 17, 2016, can provide altimeter data for continual monitoring of glaciers in this study and eleswhere.
|
|---|
