Results and Discussion

In order to understand the effect that hurricanes have on TTL moisture, we first examine lightning activity and minimum IR Tb throughout the Tropical Americas as a function of time. Figure 1 shows a map of the geographical locations used in the Hovmoller plots. Figure 2 shows the Hovmoller plots of daily lightning activity and daily minimum IR Tb from July 1 to November 30,2005. The white horizontal lines between 40° and 100° W correspond to the location of the five hurricanes analyzed in this study. First, we note the overall negative correlation between lightning frequency and IR Tb. Locations with higher lightning frequency are associated with colder cloud tops. Second, we note that both shorter-lived convection and the longer-lived hurricanes are associated with lightning activity. Furthermore, both types of convective systems have very cold cloud tops that reach deep into the upper troposphere. On average, the coldest IR Tb reached by the five hurricanes during their lifetimes was 214 +/— 16 K. Third, we note that the most intense lightning occurs mainly over water.

Lightning activity is typically associated with strong updrafts and deeper convection. This is evident in the relation with IR Tb shown in Fig. 2. Studies such as Molinari et al.'s (1998) suggested the use of lightning outbreaks in the core of hurricanes as a diagnostic for storm intensification. While our analysis is not performed on the hourly timescales and the eyewall regions used by Molinari et al. (1998), the proposed correlation is still evident on larger temporal and spatial

130°W 120°W 110°W 100°W 90°W 80°W 70 60°W 50°W 40°W 30°W Fig. 1 Geographical locations used to construct the Hovmoller diagrams shown in Fig. 2

130°W 120°W 110°W 100°W 90°W 80°W 70 60°W 50°W 40°W 30°W Fig. 1 Geographical locations used to construct the Hovmoller diagrams shown in Fig. 2

Fig. 2 Hovmoller diagrams of the number of CG lightning flashes (top) and minimum infrared brightness temperatures (bottom) averaged from 12° to 30° N over the Tropical Americas region every day. The horizontal white lines correspond to the location of the hurricanes in longitude and time. Due to the large range of values, the top panel is plotted as the exponent of log10(Lightning)

Minimum IR Tb |K) per Day

-120 -100 -80 -60 -^0 Longitude (West)

scales. Figure 3 shows time series of daily lightning activity during each of the five hurricanes analyzed in this study. The top panel shows total number of lightning flashes and minimum central pressure as a function of time as well as the number of log of Total Number of Flashes per Day log of Total Number of Flashes per Day

Fig. 2 Hovmoller diagrams of the number of CG lightning flashes (top) and minimum infrared brightness temperatures (bottom) averaged from 12° to 30° N over the Tropical Americas region every day. The horizontal white lines correspond to the location of the hurricanes in longitude and time. Due to the large range of values, the top panel is plotted as the exponent of log10(Lightning)

-120 -100 -80 -60 -40 Longitude [West)

Minimum IR Tb |K) per Day

Dennis Emily Katrins Rita Wilma

Dennis Emily Katrins Rita Wilma

Time (month/day)

Fig. 3 (Top) Time series of daily total number of flashes (black) and daily minimum storm central pressure (gray). The three black curves are for different spatial coverage: dot-dash for flashes collected within +/—1° from the storm center, dash for flashes collected within +/—3° from the storm center, and solid for flashes collected within +/—5° from the storm center. Due to the large

Time (month/day)

Fig. 3 (Top) Time series of daily total number of flashes (black) and daily minimum storm central pressure (gray). The three black curves are for different spatial coverage: dot-dash for flashes collected within +/—1° from the storm center, dash for flashes collected within +/—3° from the storm center, and solid for flashes collected within +/—5° from the storm center. Due to the large lightning flashes over several centered at +/—1°, +/— 3°, and +/—5° around the storm center. In general, this plot shows that lightning frequency increases as the storm intensifies (i.e., central pressure decreases). Hurricanes Katrina, Rita, and Wilma exhibited high lightning frequency even at their early stages. Hurricane Katrina, in particular, was unusual compared to the rest of the storms in that it maintained very high lightning frequency throughout its lifetime. Lightning frequency within +/—3° and +/—5° from the center of the storm had comparable temporal evolution and comparable magnitude at times for all hurricanes. Closer to the eye, within +/—1°, however, lightning frequency revealed a different temporal evolution. The bottom panel of Fig. 3 shows flash density at different radii from the center of the storm, namely within +/—1°, between +/—1° and +/—3°, and between +/— 3° and +/—5°. During Hurricanes Dennis, Emily, and Wilma, the distribution of flash density in all three annuli is fairly comparable in temporal evolution and in magnitude. During Hurricanes Katrina and Rita, however, flash density closer to the eye of the storm increased significantly. These findings are consistent with the findings of Shao et al. (2005). Besides areas closer to the eye of the storm, this panel also shows a significant increase in flash density over locations as far out as +/—3° from the center of these two storms. Recall we are using 2° x 2° bins of daily lightning flashes, so locations at +/—3° physically extend out an additional degree, or -100 km.

After examining the spatial and temporal distribution of lightning activity in the Tropical Americas, we proceed to examine the water vapor field below and at the bottom of the TTL, namely at 215,147, and 100 hPa. Similar to Fig. 2, we construct Hovmoller plots for MLS water vapor at the three pressure levels as shown in Fig. 4.

Some of the largest magnitudes for MLS water vapor at 147 hPa and 215 hPa are not observed exclusively over continental longitudes. Instead, both maritime and continental longitudes show significant enhancements and variability. When we examine the 100 hPa level, however, we notice that the largest magnitudes are found mostly west of Central America and without a corresponding hydration over the same longitudes at lower altitudes. This suggests easterly and upwards advec-tion of moisture to the 100 hPa level. At the hurricane longitudes, these Hovmoller plots show hydration at the 147 and 215 hPa levels usually towards the later stages of the storms. From these plots, the effect of hurricanes at 100 hPa is not entirely clear.

Next, we focus on hurricane days only and examine the moisture field and lightning frequency around the center of the storms. We construct storm-centered plots using the methodology of Ray and Rosenlof (2007). These are Lagrangian

Fig. 3. (Continued) range of values, flashes per day are plotted as the exponent of log10(number of flashes per day). (Bottom) Times series of flash density at different radii from the storm center. The three black curves are for: density within +/—1° from the storm center in dot-dash, density between +/—1° and +/—3° from the storm center in dash, and density between +/—3° and +/ —5° from the storm center in solid. Note highest density observed within +/—1° in Hurricanes Katrina and Rita (Reported degree distances from the center of the storm represent the value at the center of a 2° x 2° bin. For example, values at 3° are for measurements between 2° and 4°.)

Fig. 4 Hovmoller diagrams of MLS-water vapor at 100 hPa (left), 147 hPa (middle), and 215 hPa (right) panels averaged from 12° to 30° N every two days. The horizontal white lines correspond to the location of the hurricanes in longitude and time. Due to the large ranges in water vapor magnitudes, values at 147 and 215 hPa are plotted as the exponents of log10(H2O)

Fig. 4 Hovmoller diagrams of MLS-water vapor at 100 hPa (left), 147 hPa (middle), and 215 hPa (right) panels averaged from 12° to 30° N every two days. The horizontal white lines correspond to the location of the hurricanes in longitude and time. Due to the large ranges in water vapor magnitudes, values at 147 and 215 hPa are plotted as the exponents of log10(H2O)

plots that follow each hurricane with the origin of both x and y axes being co-located with the center of the storm. The values reported in these plots are averages of 2° x 2° bins within a given longitude and latitude range from the center of the storm for each hurricane day. Using infrared data from the Atmospheric Infrared Sounder (AIRS) instrument aboard the Aqua satellite, Ray and Rosenlof (2007) showed that tropical cyclones hydrate the 223 hPa level. Here we use microwave data at 100, 147, and 215 hPa instead. Figure 5 shows these storm-center plots for MLS water vapor and lightning frequency. These plots confirm the hydration by the hurricanes at the 147 and 215 hPa. At the 100 hPa, however, there is no evidence of direct hydration by the hurricanes. Lightning frequency, similar to 147 and 215 hPa MLS water vapor, shows increases around the center of the storm. Considering that the spatial distribution and the intensity of lightning activity varies from storm to storm, it is not surprising to find a lack of spatial correlation between lightning flashes and water vapor fields in this figure. Recall that these storm-center plots are averages over all five hurricanes.

Figure 5 also shows that the most significant hydration by the hurricanes occurs in an area that is +/— 5° in both longitude and latitude from the center of the storm. Our next step consists of exploring correlations between MLS water vapor and lightning frequency over this focused area. Since we have limited spatial and temporal

Fig. 5 Average fields of MLS-Aura water vapor and lightning frequency. The averages are calculated over daily 2° x 2° bins during all days of Hurricane Dennis, Emily, Katrina, Rita, and Wilma. The dashed inner box corresponds to the area that is +/—5° from the center of the storm

coverage from the MLS instrument, we compare only daily 2° x 2° bins within the +/— 5° area from the storm center where MLS measurements were available.

Statistical analysis on correlations among 100 hPa, 147 hPa, 215 hPa MLS water vapor, and lightning frequency reveal the existence of weak, but nonetheless statistically significant correlations at the 95% confidence level. A total of 94 data points are used for each parameter. Recall this analysis is performed while storms remain over water only. We find lightning frequency and 215 hPa MLS water vapor to have a statistically significant correlation with a linear correlation coefficient of +0.2115 and lower and upper bounds of +0.0070 and +0.3990, respectively. We also find 215 hPa and 147 hPa MLS water vapor to have a statistically significant correlation with a linear correlation coefficient of +0.2689 and lower and upper bounds of +0.0678 and +0.4490, respectively. Lastly, we find 147 hPa and 100 hPa MLS water vapor to have a statistically significant correlation with a linear correlation coefficient of —0.2936 and lower and upper bounds of —0.4701 and —0.0944, respectively. Correlation plots for these three pairs of parameters are shown in Fig. 6.

Fig. 6 Correlation plots between (top) lightning frequency and 215 hPa MLS-Aura water vapor, (middle) 147 and 215 hPa MLS-Aura water vapor, and (bottom) 147 and 100 hPa MLS-Aura water vapor. Each value is the average of water vapor or total number of lightning flashes collected in 2° x 2° bins over a day. All bins are located within +/—5° from the storm center

Fig. 6 Correlation plots between (top) lightning frequency and 215 hPa MLS-Aura water vapor, (middle) 147 and 215 hPa MLS-Aura water vapor, and (bottom) 147 and 100 hPa MLS-Aura water vapor. Each value is the average of water vapor or total number of lightning flashes collected in 2° x 2° bins over a day. All bins are located within +/—5° from the storm center

Lightning frequency was only correlated with MLS water vapor at the 215 hPa level based on statistical results. Recall from Fig. 3 how Hurricane Katrina exhibited the highest lightning frequency of all hurricanes analyzed. As shown in Fig. 6, Hurricane Katrina also had the highest 215 hPa MLS water vapor of all hurricanes. At higher altitudes, Hurricane Katrina had some of the highest water vapor and some of the lowest water vapor measurements observed at 147 and 100 hPa, respectively. At the opposite end the lightning frequency spectrum, Hurricane Emily started out with the lowest lightning frequency. This same storm also had some of the lowest observations of MLS water vapor at both 215 and 147 hPa, and some of the highest observations of MLS water vapor at 100 hPa.

Was this article helpful?

0 0
Survival Treasure

Survival Treasure

This is a collection of 3 guides all about survival. Within this collection you find the following titles: Outdoor Survival Skills, Survival Basics and The Wilderness Survival Guide.

Get My Free Ebook


Post a comment