| dc.description.abstract |
ABSTRACT
Weather radars measure the amount of mean power reflected from precipitation targets. To convert
reflectivity (Z) data into rainfall (R) data, an empirical power relationship, Z=aR
, is usually used.
However, the parameters a and b vary with time and space. These means these parameters need to
be estimated for different locations. In this thesis, new Z-R relationships are derived for Lake Tana
Basin (LTB) precipitation by applying linear regression and bulk adjustment methods using
reflectivity data from Shawra C-band weather radar and rainfall data from Bahir Dar metrological
station. By linear regression method, we found that Z=277R
1.5
. And, using bulk adjustment method
the relations, Z=270R
1.6
, Z=315R
1.5
and Z=390R
1.4
were found. These relationships are
comparable with previously derived relationships for different areas of the world in estimating the
amount of rainfall rate from reflectivity data. A correction factor between precipitation rates
measured by radar (Rr) and gauge (Rg) is also found by linear regression of these two data. The
relationship between the two precipitation rates was found to be, Rr = 1.06 Rg + 0.75. By using
this relationship, we have shown that the radar measurement captures the daily variation of rainfall
rate estimated by gauges. The mean error of rainfall estimated by the calibrated radar compared to
gauges’ measurement was 1.65 mm. This is better compared to mean errors 7.74 mm and 7.98 mm
of the new model derived by linear regression and Marshall Palmer, respectively. Further studies
about altitude profile of reflectivity and spatiotemporal pattern of precipitation over LTB in June
and July 2016 are also included. We have also presented detailed discussions and models on
weather radar signal processing including transmission, reception and filtering.
b |
en_US |