Search Results

You are looking at 1 - 10 of 24 items for

  • Author or Editor: Andrew J. Negri x
  • Refine by Access: All Content x
Clear All Modify Search
Andrew J. Negri

The cloud top structure of the Wichita Falls tornadic storm of 10 April 1979 (and other severe storms on this day) is studied using remotely-sensed observations from radar and satellite. A comprehensive data set included 3 min interval visible (0.6 μm) and infrared (11 μm) radiances from the eastern GOES and similar 30 min interval data from the western GOES. The near synchronization of these two satellites allowed for the stereoscopic determination of cloud top heights. In addition, at 2048 GMT, TIROS-N scanned the storms within one minute of the geosynchronous stereo and provided 1 km resolution infrared blackbody temperatures.

Because internal storm dynamics are hidden from the view of the satellite, storm updraft intensity must be inferred from cloud-top minimum temperature and its rate of change. The Wichita Falls, Tex. tornadic storm could be defined in the satellite data by a point of minimum temperature which displayed temporal continuity and achieved a temperature of 208 K. A cloud-top cooling rate above the tropopause of 7 K/21 min preceded tornadogenesis. An adjacent warm area (221 K) developed downwind and was surrounded by a “V”-shaped pattern of lower temperatures. The warm area is postulated as due to subsidence in the lee of an ascending tower.

The measured stereo height of the Wichita Falls storm was 15.6 km at 2349 GMT, 1.5 km higher than severe storms 150 km downwind, although its minimum blackbody temperature was 9 K higher than that of these downwind storms. In addition, unrealistic fluctuations in the time sequence of temperature 30 min prior to the Wichita Falls tornado indicate that the IR measurements are affected by sensor response and/or field of view limitations, at least close to the anvil edge. Cross sections of stereo heights, IR temperature, and radar reflectivity at 2349 GMT demonstrate that while there is, in general, a co-location of high tops, low temperatures, and high low-level radar reflectivity, significant variations can exist in height/rainfall relationships.

A comparison of data sets at 2048 GMT between stereo height measurements and IR temperatures from GOES-East and TIROS-N revealed that anvil top features can be up to 10 K warmer in the GOES field of view (100 km2) than from TIROS-N (1 km2), and that this difference can reach 20 K for young thunderstorms, with perhaps 4 K explained by calibration differences. The lapse rate for tops penetrating the tropopause was substantially closer to the adiabatic lapse rate when TIROS-N temperature minima, rather than GOES minima, were plotted as a function of stereo determined height.

Full access
Pablo Santos and Andrew J. Negri

Abstract

This paper presents a comparison of the normalized difference vegetation index (NDVI) and rainfall for the Amazon and northeastern Brazil for the time period of 1988–90. The analysis shows that the NDVI and rainfall are uncorrelated in the Amazon, except in the northernmost part, where the rainfall regime is drier and a savanna type of vegetation is present. In the drier region of northeastern Brazil, the relationship is exponential, with the NDVI showing sensitivity to rainfall within a regime of less than approximately 100 mm per month of rainfall.

Full access
Andrew J. Negri and Robert F. Adler

Abstract

Quantitative observations of thunderstorms in the midwest United States made with short-interval (5 min) geosynchronous satellite data are examined in relation to concurrent digital radar observations for one case study over a limited area. Individual thunderstorms are defined in the satellite infrared (IR) data by the location of relative minima in the equivalent blackbody temperature (TBB) field. In a large majority of cases, these satellite-defined thunderstorms coincide with individual radar echoes. This agreement allows comparison of digital satellite and radar data for individual thunderstorms.

The evolution of individual thunderstorms in terms of radar echo and satellite-observed cloud features is examined. An examination of a number of storms indicated that the first low-level radar echo (18 dBZ) appeared when the satellite observed cloud-top minimum TBB had a mean of 246 K (7.4 km). As the storms evolve, larger reflectivities appear as the cloud tops penetrate upward to colder temperatures. Larger reflectivity values (>50 dBZ) begin as the storms approach and penetrate the tropopause.

Maximum radar reflectivity is shown to be correlated with satellite-based estimates of thunderstorm intensity. Thunderstorm top ascent rates in the 235-240 K (∼8.8 km) region indicate the intensity of the initial storm updraft and are correlated with the maximum storm reflectivity with weak cells (-dTBB/dt of 1 K min−1) having maximum reflectivity of 30–40 dBZ and strong cells (3–4 K min−1) having echoes of ≥50 dBZ. The minimum TBB observed during the lifetime of the storm (Tmin), indicative of maximum storm top height, is also correlated to maximum storm rainfall. Storms with tops colder (higher) than the tropopause (212 K) have the highest rainfall rates in the severe storm situation examined here. The parameter Tmin is also very well related to maximum volume rain rate as estimated from the radar data. Storms observed to reach temperatures lower than the tropopause temperature had volume rain rates of the order 103 m3 s−1, compared to 102 m3 s−1 for weaker storms.

Full access
Robert F. Adler and Andrew J. Negri

Abstract

This paper describes a new method of estimating both tropical convective precipitation and stratiform precipitation (produced under the anvils of mature and decaying convective systems) from satellite infrared data. The method, denoted CST (Convective-stratiform Technique) locates, in an array of infrared data, all local minima in the brightness temperature field (T min. After an empirical screening to eliminate cirrus, these points are assumed to be convective centers. Rainrate and rain area are assigned to each minimum point as a function of its T min, based on one-dimensional cloud model results. A stratiform rain algorithm, using a brightness temperature threshold based on the mode temperature of thunderstorm anvils, completes the convective/stratiform rain estimation.

Individual CST rain fields wore spatially most similar to the radar for young, isolated storms, and most dissimilar in capturing linear features such as squall lines. Some convective features were missed, while others (generally cirrus debris) were sometimes misrepresented as active convection. Stratiform estimates generally corresponded to the radar-derived 1 mm h−1 contour.

The technique was tested for four south Florida cases during the second Florida Area Cumulus Experiment (FACE). Half-hourly estimates made in the FACE target area are verified against raingages and both unadjusted and gage-adjusted radar. When compared to three other infrared techniques applied to the same dataset, the CST had the lowest bias (−0.02 mm), lowest mean absolute difference (0.28 mm), lowest root mean square difference (0.39 mm), and lowest percent difference (41.2%) of any tested satellite technique.

The evolution of the precipitation averaged over the FACE target (104 km2), was well represented by the CST, particularly in capturing peak rainfall and the transition early and overestimate late, when compared to the gage-adjusted radar. Area-averaged estimates were in agreement with radar-based analyses, and comprised 10%30% of the total rainfall.

Full access
Andrew J. Negri and K. Robert Morris

Abstract

A narrow cloud-free zone of large longitudinal extent was observed in visible and infrared satellite imagery on 21 September 1978. An attempt to explain the zone in terms of subsidence induced by a transverse frontal circulation is presented.

Full access
Andrew J. Negri and Robert F. Adler

Abstract

This paper compares and contrasts three satellite infrared methods of estimating rainfall. The World Climate Research Programme's Global Precipitation Climatology Project held its first algorithm intercomparison during June–August 1989 over the Japanese archipelago and surroundings. The GOES precipitation index (GPI), the Negri–Adler–Wetzel Technique, and the convective-stratiform technique (CST) were applied to hourly infrared imagery. A network of radar data calibrated by a dense raingage network was used as ground truth but withheld from the investigators until after submission of the satellite estimates. Scattering signatures in concurrent 86-GHz brightness temperatures from the Special Sensor Microwave/Imager were used to develop a method to discriminate nonraining cirrus from active convection in two of the infrared techniques.

All three of the IR techniques did poorly in estimating the rain maxima over southeastern Japan associated with shallow orographic (warm) rain systems. The statistics for the combined 2-month dataset for both land and ocean (1.25° grid boxes) indicated that the GPI had the lowest bias (30 mm or 25% of the radar mean) but also a low correlation (.48) and high root-mean-square error (rmse) (103 mm or 87% of the mean). This was due to the GPI's overestimate in June (bias was 92 mm) and underestimate in July–August (bias was −32 mm). Despite its increased sophistication, the CST had an rmse of 104 mm, with a large negative bias (−70 mm) but a higher correlation coefficient (0.66). When the dataset was limited to the ocean-only points (to remove the effect of the shallow orographic precipitation), new statistics emerged. Under these restrictions, and for this limited dataset, the CST performed best, with the lowest bias (−39 mm or 42% of the mean), the lowest rmse (65 mm or 71% of the mean), and the highest correlation (0.79). It is believed that the lower scatter (higher correlation) of the CST and NAWT with respect to the GPI is due to the discrimination of thin cirrus used in both the NAWT and CST.

Daily rainfall estimates had rms errors of almost 200% of the mean and negative biases of about 50% of the mean. Hourly estimates for 1.25° grid boxes had rms errors of 200%–300% of the mean and negative biases of order 100% of the mean. Spatial averaging to 2.5° showed a slight improvement in these statistics. Despite the poor performance on hourly scales, the satellite techniques were able to identify diurnal signals when averaged over 1 month.

Full access
Andrew J. Negri and Thomas H. Vonder Haar

Abstract

Five-minute interval 1 km resolution SMS visible channel data were used to derive low-level wind fields by tracking small cumulus clouds on NASA's Atmospheric and Oceanographic Information Processing System (AOIPS). The satellite-derived wind fields were combined with surface mixing ratios to derive horizontal moisture convergence in the pre-storm environment of 24 April 1975. Storms began developing in an area extending from southwest Oklahoma to eastern Tennessee 2 h subsequent to the time of the derived fields. The maximum moisture convergence was computed to be 2.2 × 10−3 g kg−1 s−1 and areas of low-level convergence of moisture were in general indicative of regions of severe storm genesis. The resultant moisture convergence fields derived from two wind acts 20 min apart were spatially consistent and reflected the mesoscale forcing of ensuing storm development. Results are discussed with regard to possible limitations in quantifying the relationship between low-level flow and satellite-derived cumulus motion in an antecedent storm environment.

Full access
Andrew J. Negri and Robert F. Adler

Abstract

This study examines the relationships between satellite infrared clouds and rainfall, and infrared-threshold visible clouds and rainfall. Clouds are defined by the outline of the 253 K isotherm. Cloud infrared area was highly correlated with rain area (ρ = 0.85) and with volume rainrate (ρ = 0.81). It was poorly correlated with mean cloud rainrate (ρ = −0.28). One-parameter models were as effective in explaining the variance of cloud volume rainrate as multiparameter methods, due to the high correlations between visible brightness, mean cloud temperature and cloud area. An exception was found for clouds >10 000 km2, where area and temperature were uncorrelated, and mean temperature was more effective in discriminating among classes of volume rain than was cloud area. Statistical separation of five- of six-volume rain classes was achieved with mean temperature; however, the probability of occurrence of the classes effectively reduced this to a four-class problem.

Due to the high correlation between visible brightness and infrared temperature, visible data provided largely redundant information. Using a mean cloud brightness threshold of 148 counts, rain/no-rain separation was effected with a POD, FAR, and CSI of 0.98, 0.13, and 0.86, respectively. An infrared threshold (mean temperature of 241 K) produced statistics of 0.88, 0.07 and 0.83, respectively for the POD, FAR and CSI. The standard deviation of visible counts (used as a measure of cloud structure) was poor in explaining the variance of rainrate, yielding no better than rain/no-rain separation.

Time series of the cloud evolution showed that rain volume fluctuations were better “mirrored” by cloud temperature fluctuations than by cloud area. Contrary examples could be found and inconsistency between days was noted. The apportionment of rain volume (assigning rainrates to areas) remained a difficult problem, with significant variability, both within clouds of the same size and between clouds of different size. The coldest 10% cloud area was found to contain 11%–23% of the total rain volume while the coldest 50% area contained 60%–70–. This is in contrast to the rain apportionment used in the Griffith-Woodley Technique.

Full access
Andrew J. Negri and Robert F. Adler

Abstract

The relationships between satellite-viewed cloudy (or partly cloudy) grid cells and the variability of the precipitation contained therein are explored. Using a 32 km grid and 30 min interval visible, infrared and radar data, 5 days of the Florida Area Cumulus Experiment are examined. Cloud is delineated from no-cloud by an infrared threshold of 253 K.

While high rainrates are always associated with low temperatures, the reverse is not true: low temperatures do not always imply high rainrates. For partly cloudy cells, the percent-explained variance of rainrate by infrared parameters is low, with none of the parameters explaining more than 14% of the variance. The mean visible count explains slightly more variance, but it is not apparent that higher visible values are indicative of higher rainrates, because the higher resolution of those data introduces ground pixels into the average. When only completely cloudy cells are considered, the infrared parameters still explained about 14% of the variance, but with larger day-to-day variability. For those cells, the mean visible count explains less than 10% variance on 4 of the 5 days, due to its inability to discern rainrates in widespread cirrus anvils. The mean visible structure by itself explains 10%–26% of the rainrate variance for completely cloudy grid cells. Modest (4%–14%) increases in explained variance are shown when this quantity is then added as a second regression parameter.

Classification of the mean rainrate into six groups and the subsequent computation of a mean infrared parameter for each class shows statistically significant differences in the mean infrared parameters among classes. Assigning independent observations to classes becomes unsatisfactory given the distribution of the rain classes themselves. Variability (between days) in the mean temperature of each rainrate class is often as great as the variability (of the mean temperature) among rain classes on any given day. Relationships are clearly dependent on where in the convective cycle they occur, and this cycle is itself variable from day to day. Extensive cold anvils often produce widespread stratiform rain late in the day, while earlier these same temperatures produced intense convective rain. On the scales examined here, the results indicate that useful, accurate rainfall estimates beyond rain/no-rain discrimination are unlikely.

Full access
Andrew J. Negri, Robert F. Adler, and Peter J. Wetzel

Abstract

The Griffith-Woodley Technique (GWT) is an approach to estimating precipitation using infrared observations of clouds from geosynchronous satellites. It is examined in three ways: an analysis of the terms in the GWT equations; a case study of infrared imagery portraying convective development over Florida; and the comparison of a simplified equation set and resultant rain maps to results using the GWT. The objective is to determine the dominant factors in the calculation of GWT rain estimates.

Analysis of a single day's convection over Florida produced a number of significant insights into various terms in the GWT rainfall equations. Due to the definition of clouds by a threshold isotherm (−20°C), the majority of clouds on this day did not go through an idealized life cycle before losing their identity through merger, splitting, etc. As a result, 82% of the clouds had a defined life of 1 h (two images) or less: 64% of the defined clouds were assessed no rain because the empirically derived ratio of radar echo area to cloud area was zero for 64% of the sampled clouds. For 76% of the sample, the temperature weighting term was identically 1.0. Terms not directly related to cloud area were essentially uncorrelated with GWT rain volume, but cloud area itself was highly correlated (r=0.93). Discriminating parameters in the GWT rain apportionment algorithm were the temperatures that define the coldest 50% and coldest 10% cloud areas. Further apportionment beyond these two thresholds was found to be unnecessary. Simplifying assumptions were made to the GWT such that the resultant equations were independent of cloud life history. Application of a simple algorithm incorporating these assumptions led to daily rainfall patterns on three days that were, to first order, the same as those calculated from the GWT. Daily totals in the FACE target area were actually closer to the gage determined rain depths than the GWT estimates. Correlations between half-hourly estimates from both techniques and the gage amounts were poor. We conclude that the GWT is unnecessarily complicated for use in estimating daily rainfall. A method in which the relationship between clouds and rain is simple and straightforward can, to first order, duplicate the results of the GWT.

Full access