Search Results

You are looking at 1 - 10 of 12 items for

  • Author or Editor: G. L. Austin x
  • Refine by Access: All Content x
Clear All Modify Search
C. Warner and G. L. Austin

Abstract

Statistics of echoes observed on day 261 of GATE, using the radar aboard the Canadian ship Quadra, are described. This is a case in which detailed observations have been obtained from many sensors during the growth of a cloud cluster.

Dividing the radar field of view into many “bins,” it was found that rainfall rates and echo top heights were log-normally distributed UP to a certain limit, with a few additional high values. After identifying echo cores, bounded by a reflectivity threshold 24 dB above 1 mm6 m−3, their area were found to be distributed in a similar manner.

Upper limits of log-normality before (and during) the growth of the cluster were about 5 (and 30) mm h−1 for rainfall rate, 7 (and 7) km for echo top height (at the 29 dB level of intensity) and 100 (and 400) km2 for area (within contours of intensity 24 dBZ). To echoes organized on a length scale of about 100 km for more, one attributes maximum rainfall rates reaching roughly 50 mm h−1, peak heights reaching 14 km, and maximum areas of elongated cores reaching 5000 km2.

Full access
A. Bellon and G. L. Austin

Abstract

Empirical relationships between visible and/or IR data and rainfall rate are derived by comparing gage-calibrated radar data with colocated satellite information over Montreal, Canada. The accuracy of 1739 point gage-satellite measurements from 14 sequences of summertime rainfall during daylight hours is evaluated. The absolute difference, defined as ∑ |GISi|/∑ Gi where Gi and Si are the corresponding gage and satellite estimates, is 85%. The Critical Success Index (CSI), the Probability of Detection (POD), and the False Alarm Ratio (FAR) at the 2 mm level are of the order of 50, 70 and 35%, respectively, and the cross-correlation coefficient γ is computed to be 0.56.

The relative accuracy in the rainfall estimation of four empirical methods based on point satellite readings is determined. The scores of the visible-IR and “visible only” methods am adequate (γ = 0.56 and 0.50, respectively), but the scores of the “IR only” method are judged inadequate (γ = 0.30) because of its serious overestimation. A rain/no-rain estimate scores nearly as high (γ = 0.50) as a continuous rain estimate.

It is found that satellite estimates using our objective techniques are better than gage-interpolated estimates at locations where the nearest gage is farther than 40 km. The usefulness of these statistical satellite rainfall measurements is thus limited to the data sparse regions of the world.

Full access
S. Lovejoy and G. L. Austin

Abstract

Recent work in GOES satellite rain amount estimation techniques suggests that these satellites primarily show skill in estimating areas of rain, rather than rain amounts directly. This hypothesis is quantitatively investigated by examining the errors in estimating rain areas from the satellite data separately from the errors involved in estimating the rain amount from known rain areas. These errors, when combined, are shown to be of the same magnitude as the errors obtained by direct rain amount estimation, giving firm support for techniques using independent rain area and amount estimation schemes. For Montreal, the rms error in hourly estimates produced by the rain area estimation technique was found to be 22% as compared to the 44% for estimating the rain amount from known rain areas. For independent processes, this yields an rms error of 49% for satellite rain amount estimation.

Full access
A. Bellon and G. L. Austin

Abstract

Digital weather radar data have been used with a simple pattern recognition procedure to automatically generate precipitation forecasts in the zero to three hours range. Such a technique has been in real time operation for two years. The verification of the procedure has led to a preliminary “radar climatology” for the Montreal area in the form of a map of areas showing a predominant growth or decay of precipitation patterns.

Full access
M. Pestaina-Haynes and G. L. Austin

Abstract

Radar and synoptic data for precipitation lines from GATE, Barbados and Montreal are compared. The Montreal and Barbados lines have similar orientation but travel from opposite directions. The GATE lines have orientations almost in an east-to-west direction and the cells of the lines tend to move more nearly along their direction of elongation than the other lines. Both Montreal and GATE lines are appreciably larger than the Barbados lines. The common property of lines at the Barbados and Montreal locations is orientation of the precipitation line at a small angle counterclockwise from the steering wind. The 700, 850 and 1000 mb levels were found to be the steering levels of the Montreal, Barbados and GATE lines, respectively.

Full access
G. L. Austin and Y. D. Ahn

Abstract

A technique is described for finding an “effective pattern fallspeed” from arrays of radar data taken at different times and different heights. A weather radar whose antenna scans in three dimensions at many elevation angles inevitably takes a few minutes to record the three dimensional structure of precipitation patterns. The effective pattern fallspeed allows this rather coarse sampling time to be improved and enables the precipitation pattern to be extrapolated to the ground with the time resolution improved by a factor of about 5 to 1 min. Comparing the distance moved by a storm in this time with the spatial resolution of the radar suggests that this is close to the upper limit of useful time resolution.

Full access
B. J. Turner and G. L. Austin

Abstract

Three-dimensional radar data for three summer Florida storm are used as input to a microwave radiative transfer model. The model simulates microwave brightness observations by a 19-GHz, nadir-pointing, satellite-borne microwave radiometer.

The statistical distribution of rainfall rates for the storms studied, and therefore the optimal conversion between microwave brightness temperatures and rainfall rates, was found to be highly sensitive to the spatial resolution at which 0bservations were made. The optimum relation between the two quantities was less sensitive to the details of the vertical profile of precipitation.

Rainfall retrievals were made for a range of microwave sensor footprint sizes. From these simulations spatial sampling-error estimates were made for microwave radiometers over a range of field-of-view sizes. The necessity of matching the spatial resolution of ground truth to radiometer footprint size is emphasized. A strategy for the combined use of raingages, ground-based radar, microwave, and visible-infrared (YIS-IR) satellite sensors is discussed.

Full access
A. Bellon, S. Lovejoy, and G. L. Austin

Abstract

An algorithm yielding probability of rain from GOES visible-infrared imagery and simultaneous radar data is applied over a satellite image the size of eastern Canada. It is then mapped by means of a conic projection an a constant resolution Cartesian grid to facilitate overlay with synoptic charts. A pattern recognition technique is applied to 16 subareas of the entire map and has proved successful in tracking the displacement of the probability-of-rain contours. The potential of the system for making short-range precipitation forecasts is discussed briefly.

Full access
C. Warner, J. Simpson, G. Van Helvoirt, D. W. Martin, D. Suchman, and G. L. Austin

Abstract

Aircraft, radar, satellite and ship data, gathered on 18 September 1974 during GATE, have been brought to bear an clouds of the middle and upper troposphere associated with a cloud cluster occurring near the ridge axis of a 700 mb wave.

Clouds penetrating above 2.5 km into the middle troposphere were organized in bands about 9 km apart, aligned roughly along the direction of the wind shear in the cloud layer. Radar echoes corresponding to the cumulus congestus were of lifetime roughly 30 min, top height 6 km and peak rainfall rate 1.3 mm h−1. The number density of such echoes increased from one in 15 000 km2 to a value about three times greater, while convergence at ∼950 hPa (obtained from satellite cloud tracking) increased from about 1.5 to 3 × 10−5 s−1.

Convergence into a square box of side 150 km, circumnavigated by three aircraft in the moist layer, reached about 3.5 × 10−5 s−1 near cloud base. The upward flux of water vapor at cloud base was about 0.25 g m−2 s−1, equivalent to rain of intensity 0.9 mm h−1 with 100% conversion of the vapor. During this time mean rainfall rates over the area, and peaks averaged over 18 km2, increased from 0.2 to 0.7, and 23 to 39 mm h−1, respectively. Areas of small rainfall rate merged. High towers became taller and more numerous, but remained the same size, ∼50 km2 in area at altitude 7 km, for echo cores of intensity 29 dBZ.

A gust front at the ship Oceanographer was associated with one of the cloud bands., it was found that the band propagated discontinuously by new growth in its leading side. It featured a mesoscale pattern of updrafts in front and downdrafts behind, the downdrafts originating near altitude 2.5–3 km. A tentative conclusion is drawn that convective circulations tended to generate horizontal momentum near cloud base. The longitudinal rolls obtained theoretically by Sun (1978), for conditions of relatively strong buoyancy, match the observed bands well.

Cumulonimbus clouds reaching 15 km grew only out of an environment already moistened by lesser clouds. Cloud towers a few kilometers wide were photographed. Such towers were linked with groups of echoes identified from a high-resolution display of three-dimensional radar scans. This “wall chart” revealed that echoes were multicellular, and moved with widely differing velocities. It is deduced that individual echo groups yielded local winds of speed exceeding those in the environment. The lesser echo groups were part of a population having log-normally distributed properties. Those which yielded reflectivities as high as 46 dBZ were a different population; they were elongated norlh-northeast to south-southwest, a direction corresponding to that of a confluence asymptote at ∼950 hpa discerned from satellite data.

Full access
C. M. R. Platt, S. A. Young, R. T. Austin, G. R. Patterson, D. L. Mitchell, and S. D. Miller

Abstract

This paper presents further results on the optical properties of tropical and equatorial cirrus using the light detecting and ranging (lidar) radiometer (LIRAD) method. The results were obtained from observations in the Maritime Continent Thunderstorm Experiment (MCTEX). Values were obtained of cirrus cloud backscatter coefficient, infrared (IR) emittance, optical depth and absorption coefficient, cloud height and depth, and backscatter-to-extinction ratio. The values agree well with previous results obtained on equatorial cirrus in the Pilot Radiation Observation Experiment (PROBE) and extend those results to lower temperatures. Observations made of lidar linear depolarization ratio show similar trends to those observed in PROBE, extending those results to lower temperatures.

Regressions of cloud IR emittance and absorption coefficients are performed as a preliminary tropical dataset for both cloud-resolving and climate models. These regressions are compared with previous regressions on midlatitude and tropical synoptic cirrus clouds. The IR absorption coefficients in tropical and equatorial cirrus appear to be larger than in midlatitude cirrus for temperatures less than −40°C, with the difference increasing toward low temperatures. Thus, a significantly different relationship may be appropriate for tropical cirrus compared to midlatitude cirrus clouds.

Effective diameters of small particles in the colder tropical clouds are also measured using the ratio of visible extinction to infrared absorption. A new treatment of multiple scattering is used to correct the ratios. Effective diameters range from 6 to 9.3 μm at the colder temperatures.

Full access