Previous: Data Next: Conclusion Up: Ext. Abst.
Analysis and discussion
To investigate the relationship between El Nino and total ozone over Nairobi NINO3 data were correlated with total ozone over Nairobi. El Nino has been widely documented as significantly associated with rainfall activities over East Africa. In deed the various phases of El Nino are now been used operationally for the climate outlook over the East African region.
Kenya is characterized by two rainy seasons, namely October to December (commonly known as ‘short’) rains and the March to May (called ‘Long’) rains. Convective activity is most prominent during the long rains and much of the rain then is attributable to thunderstorm activities.
Figure 1: Temporal plot of normalized NINO3 values
An attempt to establish a relationship between stratospheric dynamics and climate variability is made by correlating total ozone with NINO3. Table 2 below shows results of the analysis.
| Jan | -0.57034 |
| Feb | -0.23762 |
| Mar | -0.41252 |
| Apr | -0.28679 |
| May | -0.18552 |
| Jun | -0.09095 |
| Jul | -0.29339 |
| Aug | 0.17345 |
| Sep | 0.406873 |
| Oct | 0.020952 |
| Nov | -0.27222 |
| Dec | -0.35314 |
Table 2: correlation between total ozone and nino3
Statistically significant relationships between NINO3 and total ozone in the first half of the year are observed. The long rains season over Kenya (March to May) suggest a negative connection with NINO3. The period before the start of the October-to-November (short) rains, on the other hand, depicts a positive relationship.
Figure 3: Temporal patterns of normalized ozone values
Figures 1 and 3 below illustrates the temporal patterns the normalized values of both total ozone and NINO3 during the long rains season in Kenya. Table 4 below shows the anomalous months/years with respect to both variables.
| Year | Month | Ozone | NINO3 |
| 1985 | ÷ (positive) | ||
| 1987 | ÷ (negative) | ÷ (Positive) | |
| 1988 | ÷ (negative) | ||
| 1992 | ÷ (negative) | ÷ (Positive) | |
| 1997 | ÷(negative) | ÷(Positive) |
Table 4: anomalous years with respect to NINO3 and total ozone.
An notable inverse relationship of the anomalous years during the long rains is a good indication of a potential use of total ozone data for prediction of extreme weather events.
Connection between Convective activity and total ozone’s temporal pattern is also investigated. METEOSAT Cold Cloud Duration (CCD) data show appreciable relationship with total ozone over the region. This is demonstrated in Table 5 and Figure 6 below:
Distinct patterns for the two rainy seasons are evident. The relationships suggest possible significant association between convective activity and possible tropospheric ozone production and/or stratospheric ozone redistribution.
| lag | correlation |
| 0 | -0.097 |
| 1 | 0.121808 |
| 2 | 0.257247 |
| 3 | 0.318821 |
| 4 | 0.325792 |
| 5 | 0.093563 |
Table 5: Lag correlation coefficient of total ozone versus thunder events
Figure 6: Temporal variablity of total ozone versus thunderstorm events in Nairobi, Kenya.