ATMO 350
Project 4: Radar-Gauge Precipitation Comparison
Due
Date: 05/05/04
Additional
information needed for project
Using WSR-88D radar data and rain gauge data,
compare the daily accumulated precipitation for Southern Arizona in July 1998.
Data
The data is presented as follows:
·
Gauge:
Gauge data only
·
Radar:
Original radar data (with no bias removal)
·
Gauge
+ (unbiased) Radar: Merged gauge and bias-adjusted radar data
Procedures
(a)
Read
the following documentation:
·
http://sgi62.wwb.noaa.gov:8080/stage4/README
·
Fulton,
et al. (see reference in the above document)
(b)
Compare
gauge and radar data using the following statistical quantities:
·
Mean
Difference (Bias)
·
Standard
Deviation of each time series
·
Standard
Deviation of the difference
·
Correlation
Coefficient
·
Statistical
Significance of the mean difference (see below)
·
Rain
Days vs. No Rain Days
(c)
Using
the radar equation and the original radar reflectivity factor (Z) from the
radar rain rate, perform the following
·
Change
the coefficient of the radar equation so the modified rain rate (denoted as R1)
is consistent with gauge data in terms of total monthly precipitation
·
Change
the exponent of the radar equation (denoted as R2) is consistent with gauge
data in terms of total monthly precipitation
(d)
Compute
the variance between R1 and gauge data, and compute the variance between R2 and
gauge data. Compare these variances. What is the method used in radar adjustment
(coefficient or exponent) in the "gau + (unbiased)" data? Which adjustment is better based on your
computation?
Statistical Quantities
The statistical significance of the mean difference
is
where x-bar and y-bar are the average of series (1)
and (2) respectively, s1 and s2 are the standard
deviations of series (1) and (2) respectively, and n1 and n2
are the number of data points in series (1) and (2) respectively
If |z| > 1.96, the difference is significant at
95% level.