Abstract
A statistical analysis of time series of area-averaged rainfall over the oceans has been conducted around the diurnal time scale. The results of our analysis can be applied directly to the problem of establishing the magnitude of expected errors to be incurred in the estimation of monthly area-averaged rain rates from low orbiting satellites. Such statistics as the mean, standard deviation, integral time scale of background red noise and spectral analyses were performed oil time series of the GOES Precipitation Index (GPI) taken at 3-hour intervals during the period spanning 19 December 1987 to 31 March 1988 over the central and eastern tropical Pacific. The analyses have been conducted on 2.5°×2.5° and 5°×5° grid boxes, separately.
The ratio of standard deviation to mean for area-averaged rain rate in the Pacific ITCZ is very regular and similar to that in GATE. Analysis of the area-averaged rainfall in the SPCZ shows a longer autocorrelation time scale than that in the ITCZ. The SPCZ exhibits significant power at the diurnal and semidiurnal frequencies, but the ITCZ shows only a marginally significant diurnal cycle in our data. The rainfall characteristics in the Pacific ITCZ appear to be similar to those in the Atlantic ITCZ in both autocorrelation time scale and diurnal variation. The mechanism driving convection in the ITCZ is suggested to be different from that in the SPCZ. The study shows that rainfall measurements by a sun-synchroneous satellite visiting a spot twice per day will include a bias due to the existence of the semidiurnal cycle in the SPCZ ranging from 5 to 10 percentage points. The bias in the ITCZ may be of the order of 5 percentage points.
