Numerical simulation of the influence of the large-scale Monsoon flow on the diurnal weather patterns over Kenya
The influence of the large-scale monsoonal wind systems on the diurnal weather patterns over Kenya was investigated in this study using a three-dimensional limited-area model known as the Regional Atmospheric Modeling System (RAMS) which was developed at Colorado State University (CSU). The contribution of the mesoscale forcings and the interaction between the mesoscale and synoptic scale systems on the weather over Kenya were also investigated. Three numerical experiments were performed in this study. Experiment 1 was devoted to investigating the influence of the large-scale monsoonal wind systems on the weather over Kenya through dry simulations of the dynamics and structure of these systems over the country. A' "variable initialization" method was applied to the European Centre for Medium-range Weather Forecasts (EC:r:l'fWF) data which was accessed from the national Center for-Atmospheric Research (NCAR) to create the initial state of the atmosphere at o Z which was nudged with another set of data at 12 Z. Starting at 0 Z the model was integrated forward in time for 24 hours. Model outputs were generated at a frequency of one hour. The second set of simulations (Experiment 2) examined the influence of the mesoscale circulations on the weather over Kenya in the absence of the large-scale monsoonal winds. The dry simulations were started from an atmosphere at rest. The initial field was procured from a one-station sounding data from Dagoretti weather station in Nairobi. A "horizontally homogeneous" technique was employed to create the initial fields for the model at 0 Z. Model integration was also performed for 24 hours to simulate the diurnal patterns of circulations that were generated by the local features in the absence of the large-scale monsoonal winds. As in the first experiment, the frequency of the model output was one hour. The model-generated mesoscale circulations (from Experiment 2) were "superimposed" on the prevailing large-scale flows at low levels (from Experiment 1) in order to examine the interaction between the mesoscale and the large-scale systems through the determination of possible areas of convergence/ divergence over the domain in the course of the 24-hour simulations. The convergence zones were taken to be areas where the mesoscale circulation was in direct opposition to the prevailing large-scale flow while the divergence zones were considered to be the regions where the mesoscale circulations and the large-scale monsoonal flows at low levels were in the same direction. The third set of simulations (Experiment 3) used the ECMWF data to generate precipitation in the convergence zones determined earlier, as explained above. This experiment also attempted to determine the control of the large-scale monsoonal winds on the space-time locations of the 'precipitation/ convergence zones over the domain. The results from the first experiment revealed the atmospheric structure over the country dominated with a shallow layer of the southeasterly monsoon winds at low levels, westerly winds at mid-levels, and easterly winds aloft. These features conformed with the observations of the case study day. The results also showed a rising motion in the low level southeasterlies (indicative of a vertical transfer of easterly momentum) including most parts of the mid-level westerlies going up to 12.0 km height, subsidence motion in the layer between 13 km and 16 km heights with an isothermal layer above to 23.0 km height. The ITez was absent at t he lowest level of the model at (J z = 122 m (::::: 990·7 illb) but appeared explicitly at CTz = 2.6 km (~741 mb) as a zone of confluence between two hemispheric air masses, namely, the southeasterly and the northeasterly trade winds. Low level pressure centers, vertical velocity, and relative vorticity fields accurately replicated regions of active weather over the domain. The East African Low Level Jet Stream (EALLJ) at the coast and the Turkana Jet Stream channelled in the Turkana-Marsabit corridor were also revealed in the simulations. The large-scale moisture field at low levels (O'z = 122 m) showed a maxima of 22.4 g kg-1 at the coastal region at 0600 UTe which spread inland advected by the large-scale winds in the course of the simulations. The results from the second experiment signified the importance of the mesoscale circulations on the space-time characteristics of weather over Kenya. The results clearly identified the contributions of the local circulations comprising the sea/land and lake/land breeze circulations plus the upslope/downslope drainage winds. The dry simulations further helped to reveal the dynamics and structure of the atmosphere in the absence of the large-scale flow fields. The moisture maxima (19.2 g kg-I) recorded at the coastal region (at O'z = 122 m) spread inland advected by the sea breeze coupled with the upslope winds. The results from the third experiment showed the space-time location of the precipitation generated in the convergence zones which were controlled and advected by the large-scale flow field. The results of this experiment revealed the diurnal characteristic patterns of weather over Kenya. The coastal region, the eastern slopes of the Kenyan Highlands, the Kenyan Highlands, and the Lake Victoria trough region depicted diurnal weather patterns that were related to the interactions between the local circulations such as the sea/land breeze circulations, upslope/downslope drainage winds with the prevailing southeast monsoon winds at low levels. The model-generated results showed that although the local features like Lake Victoria, the Indian Ocean, orographic barriers, Turkana-Marsabit corridor, etc. have a strong influence on the local circulation patterns of weather, the large-scale monsoon winds contribute substantially to the inland transport of large amounts of moisture into Kenya. Lastly, it was concluded from the results of this study that mesoscale and synopticscale weather systems together with their scale-interaction playa crucial "role in the de.:',. termination of the space-time characteristics of the weather observed over Kenya. Since no operational limited-area model is currently available in Kenya, this work gives both an insight as well as forming a major basis for the development or adoption of an operational limited-area model for Kenya. This will facilitate forecasts of the rainfall events over the region which may include thunderstorms, hailstorms, lightning, dry/wet spells, strong winds, risk zones for the weather hazards, etc. Such informations are crucial to the development, planning, and mitigation of all weather dependent activities, and will help the government to minimize the severe socio-economical implications of the adverse weather events.