Daily Rainfall Totals Research Articles

Extreme value theory for processes with periodic variances

Suppose {Xn n ≥ 1} is a stationary sequence satisfying the D and D1 mixing conditions given by Adler [1]. Suppose further that g and h are functions on the positive integers such that h is positive, periodic with integer period p ≥ 1 and g satisfies a certain growth rate condition. By first exhibiting the weak convergence of a sequence of point processes related to , we derive the asymptotic distribution of . When {Xn} is in the maximal domain of attraction of A(x) = exp(-e -xx), {Mn} is attracted to (A(x))q/p where . If {Xn} is in the maximal domain of attraction of G(x) = exp(-x -α) then {Mn} is attracted to (G(x)) θ where . These asymptotic results lead to a new method for estimating the level of the k-year return period, a quantity of interest in many climatological studies. An analysis of the daily rainfall totals in Homestead, Florida from 1939-1979 is provided, illustrating this estimation procedure.

Storm rainfall on the Njemps flats, baringo district, Kenya

AbstractAnnual, seasonal and storm rainfall characteristics are analysed for the Njemps Flats, a semi‐arid area near Lake Baringo in the Kenya Rift Valley. Storm rainfall is analysed in terms of intensity, timing, duration and total storm depth. Storms characteristically occur in the early evening, with one storm of around 2 h making up the daily rainfall totals. High intensity rainfall seldom persists longer than 30 min. The maximum 30 min intensity recorded was 80 mm/h, the maximum 60 min intensity was 75 mm/h. An analysis of annual and monthly recurrence intervals of storm depth indicate that for a 1 in 10 year recurrence interval, June, July and April suffer the heaviest falls, whereas for a 1 in 2 year recurrence interval the months with the heaviest storms become August, July and April. The significance of this to erosion potential and runoff harvesting are discussed.

Extreme value theory for processes with periodic variances

Suppose {Xn n ≥ 1} is a stationary sequence satisfying the D and D1 mixing conditions given by Adler [1]. Suppose further that g and h are functions on the positive integers such that h is positive, periodic with integer period p ≥ 1 and g satisfies a certain growth rate condition. By first exhibiting the weak convergence of a sequence of point processes related to , we derive the asymptotic distribution of . When {Xn} is in the maximal domain of attraction of A(x) = exp(-e -xx), {Mn} is attracted to (A(x))q/p where . If {Xn} is in the maximal domain of attraction of G(x) = exp(-x -α) then {Mn} is attracted to (G(x)) θ where . These asymptotic results lead to a new method for estimating the level of the k-year return period, a quantity of interest in many climatological studies. An analysis of the daily rainfall totals in Homestead, Florida from 1939-1979 is provided, illustrating this estimation procedure.

Non-infectious factors affecting the presentation of dairy cows to artificial insemination and the result of mating in south western Victoria.

The records kept by 4 artificial breeding centres in south western Victoria were examined and analysed by multiple regression analysis to determine whether daily fluctuations in the number of dairy cows presented for insemination, and daily, monthly and annual fluctuations in the result of first service, were associated with either climatic conditions or nutritional factors assessed as the availability of conserved hay and oats as feed supplements in the district. Over a 122 day observation period from July to October 1974, the number of cows presented daily for insemination was negatively correlated with daily rainfall total (p less than 0.01), wind speed (p less than 0.001) and barometric pressure (p less than 0.05). When the mean value for barometric pressure was used in the regression equation, rainfall totals and wind speeds above zero were associated with a mean reduction of 13.8% in the number of cows presented daily for insemination. The daily percentages of cows not resubmitted for service within 60 days of first service (non-return rates) were unrelated to climatic conditions and to the number of cows presented for insemination. The monthly non-return rate had a negative correlation (p less than 0.001) with the cost of oats but was not related to corresponding climatic changes. Annual non-return rate fluctuations were negatively correlated with the cost of oats (p less than 0.001) and with the July to October rainfall total (p less than 0.01). It is suggested that the correlation of non-return rate with the July to October total rainfall was due to an effect of rainfall total on the nutritional value of the pasture.

Spatial interpolation of daily totals of rainfall

Some experiments concerning the spatial interpolation of daily rainfall totals are reported. Several well-known techniques were evaluated in an experimental way, namely by using the given technique for the estimation of the rainfall amount at a place where the rainfall was also measured. In each experiment the same data set was used, so comparison of the performance of the different techniques was quite simple. The used techniques were all of the type of weighted average of surrounding stations (which type includes also plane-fitting techniques but not — for instance — spline fitting, etc.). The results of these experiments indicate that for the western part of The Netherlands the choice of weights is not very critical and that optimization of weights is hardly necessary. This paper is a summary of a Research Report of the Royal Netherlands Meteorological Institute.

Estimation and dimulation of sheet run-off

Sheet run-off was estimated for a Townsville stylo-annual grass pasture with 1% slope in the northern Australian monsoonal region by using neutron hydrometry supplemented by gravimetric soil sampling, both carefully timed with respect to the rainfall event. The technique is applicable to run-off and run-on situations. An empirical model was developed and incorporated into a water balance routine in an existing growth model. It relates surface sheet run-off to the soil antecedent water content (top 20 cm) and the standard daily rainfall total; the coefficients of the empirical functions are site specific. For the rainfall events studied, run-off was 37 � 12% of the rainfall. The soil water values simulated with the run-off model included were significantly closer to the experimental values than when run-off was assumed to be zero. This improvement in simulation of soil water profiles was consistent with an improvement in subsequent dry matter simulation.

Rainstorm Distributions in Tanzania

Rainstorms, short periods of rather intensive rainfall, constitute a much larger proportion of total rainfall in the tropics than in the mid-latitudes. Rainstorms strongly influence both the variability in time and the spatial distribution of tropical rainfall. The most important effects are due to the high rainfall intensity during these storms, which can result in serious damage in the form of floods, soil erosion or landslides, and the washing away of agricultural crops. However, rainstorms have largely been disregarded as a factor in regional and agricultural planning because of the lack of suitable basic data. Rainstorms in Tanzania are studied on the basis of daily rainfall totals, which are available for many stations and over sufficiently long periods. It is shown that these data are good indicators of the main characteristics of rainstorms. The seasonal distribution shows little difference from that of the mean rainfall. The frequency of occurrence, contribution to total rainfall and the mean rainfall intensity for all days with more than 25 mm precipitation are expressed in annual indices. These are used in the construction of quantitative maps, which illustrate the distribution of these features. The maps identify areas where the threat of rainstorms is greatest. As the development of rainstorms is largely controlled by local factors, and since the storms frequently move rather erratically, the exact location where they will strike cannot be predicted and the maps are therefore mainly indicators of probabilities.

Cloud Seeding Effects on Different Daily Rainfall Amounts

Abstract The results of the cloud seeding experiment in New England, Australia, have been stratified to show different effects of seeding on different daily rainfall totals. No effect was detected on days with an area average rainfall of less than 0.1 inch, which contributed about 21% to the total rainfall. Days with between 0.1 and 0.5 inch, contributing 45% to the total rain, showed increases of 10–20%. The 0.5–1.0 inch class, contributing about 22%, showed an apparent 10% decrease. There were too few cases to give a meaningful result for heavier falls.

THE FREQUENCY OF SMALL DAILY RAINFALL AMOUNTS IN EASTERN UNITED STATES

Abstract Summary This paper presents a simple procedure for estimating the frequency of daily rainfall totals up to 1-inch in magnitude for the United States east of 100° W. Realistic results can be obtained from three relationships presented in the form of diagrams and the two generally available statistics: mean annual precipitation and mean annual number of days of precipitation greater than or equal to .01 inch. The procedure developed here gives reasonably accurate results, particularly for the rainfalls which occur frequently. A logical question is then one of extensions to other regions and refinements. This procedure can probably be extended to other regions by using thresholds other than the .01 and the 1 inch considered here. A combination of more stations with records longer than the 20 years utilized in this study will undoubtedly improve the quality of the estimates.

Meteoric dust and rainfall

In a recent communication Bowen (1953) has drawn attention to the fact that on certain days heavy fall of rain occur year after year in different parts of the world and if total daily rainfall values of one of these places for a fairly long number of years are plotted on a graph then one would get peaks of heavy rain on those days.