National Aeronautics and Space Administration

Clouds Generated During Midlatitude Storms

By Kwajo Abeyie

Abstract

This research project examines the cloud structures produced during storm lifecycles. To study the clouds in detail we use a variety of data sets including records from ground based observations collected by the National Centers for Environmental Prediction (NCEP) formerly known as the National Meteorological Center (NMC) and data from satellite observations collected by the International Satellite Cloud Climatology Program (ISCCP). Through the use of various FORTRAN programs, the NMC data set enables us to identify storms and then follow the paths these storm take based upon their mean Sea Level Pressure (SLP). The ISCCP data set then allows us to examine the properties of the clouds associated with each of these storms. In examining the ISCCP data for a storm on January 8, 1988 we found that not all the clouds formed by storms follow the classical viewpoint of distribution along frontal patterns. For example, we find the greatest concentration of deep convective clouds to be along the warm front and not along the cold front as would have been expected in the classical description of cloud type distribution in storms. A similar discrepancy was also found with the nimbostratus clouds which formed along the cold front and not along the warm front as we would have expected. Possible explanations for these discrepancies are examined.

The actual path Storm 9 took in January 1988 (the eighth to fifteenth days).

The extraction box for the first observation of Storm 9.

Sea Level Pressure of Storm 9 throughout its duration.

Introduction

The purpose of this research is to contribute to the understanding of how cloud structures develop during storm life cycles. Storms are studied in the mid-latitude range, 30 to 60 degrees, in the northern hemisphere. This location was selected because this is where most of the Earth's land masses are and a majority of the population lives there. The primary purpose of examining storms is to determine their effect on weather and climate.

The most important role clouds play in the climate process is their role in the basic radiation balance of the earth. Clouds reflect about 30 and 60 percent of the sunlight striking them, giving clouds their bright, white appearance (Rossow, 1993). A cloudless earth would absorb nearly 20 percent more heat from the sun than the present day earth does (Rossow, 1993). Clouds form when surface water evaporates, creating humidity, and then condense on microscopic airborne particles. The cloud formation process is part of the hydrologic cycle. This cycle traces the path of water as it goes through various phase changes. When water is evaporated from bodies of water, the phase change to vapor allows it to rise into the atmosphere. After the rain falls the water soaks back into the earth or runs back into the sea where it can once again be evaporated (Gedzelman, 1980).

Clouds not only reflect incoming sunlight, increasing the planetary albedo and thereby creating a cooling effect but also let light pass through, creating a warmer surface environment below them. This is due to the optical thickness of the cloud. Cloud optical thickness is determined by a process called remote sensing. Sunlight is reflected off the tops of clouds into space and is captured by orbiting satellites. This information is then used to help determine what types of clouds are present. If the light reflected is less intense than the incoming sunlight, the cloud is then classified as a thin cloud. Clouds located high up in the atmosphere, called cirrus clouds, are examples of thin clouds (Harvey, 1994). However, if light reflected is greater than the incoming sunlight, the cloud is classified as optically thick. An example of an optically thick cloud would be a nimbostratus cloud that contains rain and is dark in appearance from the surface. A nimbostratus cloud would appear very bright to a satellite because a lot of light is being reflected off the tops of the clouds (Harvey, 1994).

In addition to the reflective effect of clouds, clouds can also influence the greenhouse effect (which is produced when carbon dioxide and other atmospheric gases are released into the earths atmosphere then traps it as heat creating a warming effect). Since clouds contain water droplets they tend to act as a barrier to heat trying to escape into space. In essence, condensed water droplets act similarly to greenhouse gases. Without the greenhouse effect, Earth would be too cold for life to exist (Briggs, 1995).

Hypothesis

In mid-latitude storms, deep convective clouds form along warm fronts because this system creates a vertical cloud structure.

Research Approach

1. In order to find clouds we need to find storms, because storms are the major producers of clouds.
   
   
2. Low sea level pressure is a good indicator of a storm, because storms are simply a series of low pressure systems.
   
   
3. Low pressures are located using a FORTRAN computer program.
   1. How the program locates the storm centers: Currently, three meteorological parameters are available as the indicators for selecting storms. However, only the actual sea level pressure was utilized to simplify the investigations.

      The present search strategy is to use the three indicators independently for searching storms. In the future, there can be a combination of the different indicators for searching storms. A storm is defined as a grid box whose pressure (or vorticity) is lower than its eight neighboring grid boxes by a prescribed threshold (usually 1 millibar). If any neighboring grid boxes does not meet the requirement, then the next three outer neighboring grid boxes will be used to check if the differential pressure (or vorticity) meets the requirement. The idea is to find a closed circle around a grid box whose pressure (or vorticity) is lower (or higher) than its neighboring grid boxes. The following diagram illustrates the search strategy.

      A15	A25	A35	A45	A55
      A14	B24	B34	B44	A54
      A13	B23	C	B43	A53
      A12	B22	B32	B42	B52
      A11	A21	A31	A41	A51

      
      

      In this diagram, C is the center, B's are the eight neighboring grid boxes, and A's are the next tier of neighboring grid boxes. C is considered a storm center if the pressure (or vorticity) of C is lower (or higher) than all B's by a prescribed threshold. If any B-grid box does not meet the requirement, for example, B32, then the three outer neighboring grid boxes, that is A21, A31 and A41, will be checked if they meet the requirement. If B22 does not meet the requirement, then A12, A11 and A21 will be checked. Once a closed circle can be found around a grid box, that grid box is considered a storm center.

   2. How the program tracks the storm throughout its life cycle: Storm tracking can be performed with either sea level pressure or sea level pressure anomaly. The following criteria are used in tracking storms:
      1. The highest mean speed of propagation of a storm in a twelve hour period is approximately 140 km/hour.
      2. A storm is not allowed to exhibit motion, that is westward propagating, in a twelve hour period.
      3. Storms are not allowed to split. This means that if there are more than one storm centers within the allowed range of a center from the previous time step, the storm center with the lowest pressure (or pressure anomaly) is assumed to be associated with the center from the previous time step.
      4. Storms are not allowed to merge. This means that if there is more than one storm center within the allowed range of a center from the subsequent time step, the storm center with the lowest pressure (or pressure anomaly) is assumed to be associated with the center from the subsequent time step.
      5. A storm has to exist for at least five consecutive time steps in order to qualify as a storm. (One time step is 12 hours).
   
   
4. A second FORTRAN program examines the sea level pressure around the storm center. This program determines the size of the storm area by working outwards from the central pressure until the pressure decreases again. An extractions then made from that whole central pressure. This extraction area is then used with the ISCCP data set to examine the properties of clouds within that storm.

   Storms are then categorized by several different methods: For example, storms can be classified according to their duration in terms of the number of twelve hour intervals it has. An average length is taken of all the storms given and then each storm is categorized as below average, average, and above average. Below Average Storms are the storms that last less than seven consecutive twelve hour observations.

   Average Storms are the storms that continue for seven, eight, or nine consecutive twelve hour observations. Above Average Storms are those storms that endure for more than nine consecutive twelve hour observations.

   Categorizing storms according to their sea level pressure patterns is another way of classifying storms. By plotting central sea level pressure against time a pressure profile can be created. All storms do not conform to the same sea level pressure pattern throughout its life. Some storms can be classified as classical storms. These are storms where their pressure profiles are "V" shaped or have a dip pattern in pressure. Oscillating storms are those with insignificant pressure changes, while ascending storms increase in pressure over time. Descending storms are those that decrease over time in pressure only.

5. A third method is to plot the actual tracks of these storms as they move across the earth. The categories created through the storms pressure profiles and the storm tracking program can then be compared to see if there is any correlations. After making these examinations, several storms are selected for more detailed analysis. A third computer program examines the ISCCP data from the observations of the selected storms and produces histograms of the cloud properties are plotted to identify different cloud types. The story of a storm can then be told by examining how its cloud types change throughout its life and how these changes relate to other variations in the storms property.

   Comparisons between the various classifications were made to observe any correlations between storm duration, pattern (classical, oscillating, ascending, descending), or location. When comparing the storms pressure profiles and their locations, no correlations were found among the classical storms, ascending storms, or the descending storms. However, there was a correlations between the oscillating storms and their location. They were usually located over land in Asia, in the region of the Himalayan mountains. This is probably due to the NMC being doubtful at high altitudes. Comparisons made between storms duration and location showed no correlation.

Results

The main finding was that deep convective clouds formed along a cold front system that conflicts with a classical view (a source). According to the classical view point, deep convective clouds form along warm front systems (to creatucture).

The sea level pressure of Storm 9 at its first observation at hour twelve.

The wind direction at the 700mb level. It shows an easterly flow (red displays an easterly flow of wind as the blue displays a westerly wind flow).	This displays a counter clockwise wind flow around the central low pressure horizontally (red displays an easterly flow of wind as the blue displays a westerly wind flow).
This displays a counter clockwise wind flow around the central low pressure vertically (blue shows a southerly wind flow as the red shows a northerly wind flow).	The winds are flowing in a South to North direction (blue shows a southerly wind flow as the red shows a northerly wind flow).
This is the height of the cloud tops in millibars.	This is the optical thickness of all cloud types (between the first and second observations).
The relative humidity is abundant around the low pressure.	Warm and cold air masses meeting and establishing fronts

ISCCP Cloud Classification Chart

Cirrus clouds forming mostly along the cold front.	Cirrostratus clouds are forming along the warm front.	The deep convective clouds are abundant along the warm front.
A few altocumulus clouds form along the cold front.	Most of the altostratus clouds are formed along the cold front.	A few nimbostratus clouds are formed along the cold front.
A few cumulus clouds form along the cold front.	Most of the stratocumulus clouds formed along the cold front.	The distribution of the stratus clouds are formed along the cold front.

Conclusion

There are several possible explanations for the discrepancies that were observed. The source used for the classical viewpoint is only a generalization of where clouds form along fronts. The radar device used to obtain different cloud structures in storms by this source were unable to determine cloud height accurately. Thinner clouds were near the center of the low pressure and thicker clouds behind the low pressure as the storm progressed. Storms may have different characteristics at different latitudes due to differences in the mean SLP. A change in latitude also results in a change in temperature (Gedzelman, 1980). Cloud data was missing for many extractions due to the inability to obtain data during the night time. This missing data could have accounted for these findings.

Future Research

These results along with other results will act as a foundation for future research and the current understanding of cloud structures in storm life cycles. For further investigations, the same process will be used for different months with their storm tracks to further validate the findings.

Acknowledgements

The FORTRAN programs used to analyze cloud data were written by Ken Lo, Audrey Wolf, and Robert Kruckeberg, at the NASA Goddard Institute for Space Studies.

References

Briggs, A., D. Blaustein, C. Kapicka, A. Kaskel, and L. Lundgren. 1995. Biology: The Dynamics of Life. Glencoe/McGraw-Hill.

Gedzelman, S.D. 1980. The Science and Wonders of the Atmosphere. John Wiley & Sons, Inc.

Harvey, D. 1994. Kids Discover.

Rossow, W.B. 1991. Who knows where the clouds go? The Sciences, May/June, 36-41.

Wallace, J.M., and P.V. Hobbs. 1977. Atmospheric Science: An Introductory Survey. Academic Press, Inc.

This poster was written by Kwajo Abeyie in August 1996, a student at A. Philip Randolph High School at the time. GISS science advisors on the project were Drs. Anthony Del Genio, George Tselioudis, and William Rossow.