<br />AUGUST 1984
<br />
<br />DONEAUD, IONESCU-NISCOV AND MILLER,JR.
<br />
<br />1611
<br />
<br />volumes using satellite data. Many of the satellite
<br />schemes (Augustine et al., 1980; Griffith et al., 1981;
<br />Stout et aI., 1979) developed toestimate rain amounts
<br />depend on the delineation of both the rain area and
<br />the lifetime of the storm. Larger errors in delineating
<br />rain areas from satellite pictures might be generated
<br />by the stratification of the convective components
<br />and other storm-weakening processes during its de-
<br />caying phase. A scheme based only on data from the
<br />growing period of the convective cloud history may
<br />help reduce errors now encountered.
<br />
<br />..
<br />
<br />7. Conclusions
<br />
<br />Radar,data from the 1980 and 1981 NDCMP were
<br />used to investigate the evolution of rain rates during
<br />storms in the semiarid climate of North Dakota. The
<br />conclusions are:
<br />
<br />"'1',"
<br />
<br />1) The value of the average rain rate over the
<br />storm duration depends primarily on the reflectivity
<br />threshold considered in calculating the area-time
<br />integr'al (the area coverage integrated over the lifetime
<br />of the storm). For a selected threshold, the average
<br />rain rate R could be considered in a ,first approxi-
<br />mation independent of the A TI. The 25 dBz A TI
<br />threshold seems to be a suitable compromise, but the
<br />very small. clusters should be ignored.
<br />2) When a 25 dBz A TI threshold was used, the
<br />average rain rate is -4 mm h-1 for a dry season (i.e.,
<br />1980) and -4.8 mm h-I for a wet season (i.e., 1981)
<br />. in North Dakota. This suggests that the average rain
<br />rate depends on weather conditions, being - 20%
<br />higher for wet conditions. In Florida, the average rain
<br />rate was found to be 2.5 times larger and was
<br />considered independent of the rain volume.
<br />3) The division of a cluster or storm into its
<br />growing and decaying periods was made by consid-
<br />ering the radar scan with maximum 'echo area (for
<br />A TI ~ 25 dBz), maximum reflectivity, or radar
<br />maximum echo height. On average, a cluster reached
<br />its maximum growth after - 56% of the total cluster
<br />lifetime.
<br />4) The average rain rate for the growing period
<br />exceeds that for the decaying period by an average of
<br />-20%; in comparison, in the subtropical climate of
<br />south Florida the rain rate for the growing period
<br />was found to be twice that for the decaying period.
<br />5) When the single-scan, area-time product was
<br />used in the rain rate computation, the scatter of the
<br />data increased. The correlation coefficient dropped
<br />from 0.98 to 0.94 and the standard error of estimate
<br />increased from 0.14 to 0.20. As the time increment
<br />used in the A TI computations comes closer to the
<br />total storm duration, the scatter of the average rain
<br />rates is reduced and the predictive power of the A TI
<br />increases. A comparison of the rain volumes computed
<br />
<br />~
<br />
<br />..
<br />
<br />
<br />
<br />from the 1981 radar reflectivity data using both the
<br />1980 average rain rate and the standard Marshall-
<br />Palmer relationship applied to the 1981 data showed
<br />no significant differences.
<br />6) A multiple linear regression analysis demon-
<br />strated that the radar-estimated rain volqme is well
<br />correlated with the maximum single-scan rain volume,
<br />suggesting the possibility that the total rain volume
<br />of a storm can be estimated following identification
<br />of its maximum stage of development. This could
<br />improve satellite rain volume estimates since larger
<br />errors might be encountered in such calculations. due
<br />to overestimation of rain volumes during a storm's
<br />weakening or decaying phase.
<br />
<br />Acknowledgments. Support for this research was
<br />partially provided by the North Dakota Weather
<br />Modification Board under Contract WMB-IAS-80-1
<br />and partially by the National Aeronautics and Space
<br />Administration under Grant NAG-5-386. The paper
<br />is based in part on a thesis submitted by Stefano
<br />Ionescu-Niscov in fulfillment of the requirements for
<br />the M.S. degree in meteorology.
<br />Thanks are given to Mr. Dave Priegnitz for sub-
<br />stantial efforts in programming and computer pro-
<br />cessing of the radar data tapes, to Dr. P. L. Smith for
<br />discussing the manuscript, and to the unknown re-
<br />viewers for their helpful suggestions. Special thanks
<br />are given to Sandra Palmer and Joie Robinson for
<br />work on the manuscript, and to the late Mel Flan-
<br />nagan for drafting the figures.
<br />
<br />REFERENCES
<br />
<br />Augustine, J. A., C. G. Griffith, W. L. Woodley and J. G. Meitin,
<br />1980: Insights into errors of SMS-inferredGATE convective
<br />rainfall. J. Appl.Meteor., 20, 509-520.
<br />Austin, P. M., and R. A. Houze, Jr., 1972: Analysis of the structure
<br />of precipitation patterns in New England. J. Appl. Meteor.,
<br />11, 926-935.
<br />Byers, H. R., 1948: The use of radar in determining the amount
<br />ofrain falling over a small area. Trans. Amer. Geophys. Union,
<br />29, 187-196.
<br />Doneaud, A. A., P. L. Smith, A. S. Dennis and S. Sengupta, 1981:
<br />A simple method for estimating convective rain volume over
<br />an area. Water Resour. Res., 17, 1676-1682.
<br />-, S. Ionescu-Niscov, D. L. Priegnitz and P. L. Smith, 1984:
<br />Thl~ area-time integral as an indicator for convective rain
<br />volumes. J. Climate Appl. Meteor., 23, 555-561.
<br />Griffith, C. G., W. L. Woodley, S. Browner, J. Teijeiro, M. Maier,
<br />D. W. Martin, J. Stout and D. N. Sikdar, 1976: Rainfall
<br />estimation from geosynchronous satellite imagery during day-
<br />light hours. NOAA Tech. Rep. ERL 356-WMPO 7, 106 pp.
<br />[Available from Environmental Research Laboratory, NOAA,
<br />Boulder, CO 80303.]
<br />-, --, P. G. Grube, D. W. Martin, J. Stout and D. N. Sikdar,
<br />1978: Rain estimation from geosyn<;hronous satellite imagery-
<br />Visible and infrared studies. Mon. Wea. Rev., 106, 1153-1171.
<br />-, J. A. Augustine and W. L. Woodley, 1981: Satellite rain
<br />estimation in the U.S. High Plains. J. Appl. Meteor.. 20, 53-
<br />66.
<br />
|