Insights into Errors of SMS-Inferred GATE Convective Rainfall

Augustine, John; Griffith, Cecilia Girz; Woodley, William L.; Meitín, José G.

doi:10.1175/1520-0450(1981)020<0509:iieosi>2.0.co;2

Cited by 14 publications

(1 citation statement)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the technique which uses infrared data, a threshold value (usually -20* of brightness temperature) is used as rain-cloud identification. However, some erroneous estimations of rainfall amount due to the long-lived anvil and to low level clouds were reported (Stout et al, 1979;Augustine et al, 1981;Negri and Adler, 1984;Wash et al, 1984). This implies shortcomings of the rain area delineation using the infrared threshold technique.…”

Section: Introductionmentioning

confidence: 99%

An Instantaneous Delineation of Convective Rainfall Areas Using Split Window Data of NOAA-7 AVHRR

Inoue¹

1987

Journal of the Meteorological Society of Japan

View full text Add to dashboard Cite

A feasibility study of delineation of convective rainfall area is presented using the split window (11 µm and 12 µm) data measured by the Advanced Very High Resolution Radiometer (AVHRR) on board NOAA-7. Non-precipitating cirrus clouds and low-level cumulus clouds, which cause a erroneous rainfall area delineation, and cumulonimbus clouds are objectively distinguished by the split window data. The satellite rainfall areas, in this study, are defined as those of cumulus and cumulonimbus clouds whose tops are over 700mb level. The satellite rain-clouds are compared with concurrent radar data for eight cases. The study was performed over the area of 33°-39°N and 135°-14*E during the summer season in 1984 by taking into account the availability of digital radar data and the atmospheric condition for convective clouds.Statistics show that this method is better than the conventional single infrared method (utilizing -20* of brightness temperature as a threshold for rain clouds) by 12% for the probability of detection and 13% for the false alarm ratio. The false alarm ratio by our method is generally better than the conventional single infrared method for any cases studied. The improvement of the false alarm ratio by our method over the single infrared method is comparable to that shown by other authors who use the method combining visible and infrared data. Significant improvement in rain area delineation by our method in comparison with the single infrared method can be obtained in the situation where single layer cirrus clouds or low-level rain-clouds dominate over the study area.

show abstract

Section: Introductionmentioning

confidence: 99%