Skip to Main Content

Instead, alternative estimates were derived based on this smaller number of raingauges and the net inflow record itself. Table 4 summarises the approaches that were used which were as follows:

Table 4

Summary of lake rainfall and rainfall index series discussed in the text

SeriesHydrological year (Nov–Oct)TypeBasis of approach
WMO (1983) lake rainfall 1954/55–1979/80 Monthly lake rainfall estimates Weighted average of 17 raingauge records of which 12 were around the lakeshore in Malawi and 4 along the Tanzanian lakeshore, with the remaining gauge on Likoma Island in the Malawi part of the lake. In the weighting scheme used, the gauge records from Malawi accounted for about 80% of the total 
WMO (1983) climate index 1920/21–1979/80 Annual index series Weighted average of 10 raingauge records, all from Malawi, of which 4 were used in the above estimation procedure and the remainder were of necessity from locations more distant from the lake, but within or near the lake catchment. Approximately two-thirds of the contribution to total values was from the following 4 gauges: Nkhota Kota, Livingstonia, Karonga and Chinteche 
WMO (1983) climate index (present study) 1920/21–1979/80 Monthly index series The annual WMO (1983) values disaggregated to monthly values using a seasonal profile. The profile for the Nkhota Kota gauge was used since a comparison with the WMO (1983) lake rainfall series showed this to be the most representative record, when compared with those for the Mzimba and Kasungu gauges. To help with infilling missing periods in the lake rainfall, the profile for the period to 1953/54 was used 
Raingauge regression model (present study) 1933/34–2008/09 Monthly index series A fixed parameter multiple regression relationship developed between the scaled logarithms of the Nkhota Kota and Mzimba records and the WMO (1983) lake rainfall record 
Net inflow regression model (present study) 1899/00–2008/09 Monthly index series A fixed parameter linear regression relationship between the net inflow record and the logarithm of the WMO (1983) lake rainfall record, with any negative estimated rainfall values set to zero for the purpose of this approximate analysis; the net effect of this assumption was to change the mean lake rainfall estimate by about 2–3% 
SeriesHydrological year (Nov–Oct)TypeBasis of approach
WMO (1983) lake rainfall 1954/55–1979/80 Monthly lake rainfall estimates Weighted average of 17 raingauge records of which 12 were around the lakeshore in Malawi and 4 along the Tanzanian lakeshore, with the remaining gauge on Likoma Island in the Malawi part of the lake. In the weighting scheme used, the gauge records from Malawi accounted for about 80% of the total 
WMO (1983) climate index 1920/21–1979/80 Annual index series Weighted average of 10 raingauge records, all from Malawi, of which 4 were used in the above estimation procedure and the remainder were of necessity from locations more distant from the lake, but within or near the lake catchment. Approximately two-thirds of the contribution to total values was from the following 4 gauges: Nkhota Kota, Livingstonia, Karonga and Chinteche 
WMO (1983) climate index (present study) 1920/21–1979/80 Monthly index series The annual WMO (1983) values disaggregated to monthly values using a seasonal profile. The profile for the Nkhota Kota gauge was used since a comparison with the WMO (1983) lake rainfall series showed this to be the most representative record, when compared with those for the Mzimba and Kasungu gauges. To help with infilling missing periods in the lake rainfall, the profile for the period to 1953/54 was used 
Raingauge regression model (present study) 1933/34–2008/09 Monthly index series A fixed parameter multiple regression relationship developed between the scaled logarithms of the Nkhota Kota and Mzimba records and the WMO (1983) lake rainfall record 
Net inflow regression model (present study) 1899/00–2008/09 Monthly index series A fixed parameter linear regression relationship between the net inflow record and the logarithm of the WMO (1983) lake rainfall record, with any negative estimated rainfall values set to zero for the purpose of this approximate analysis; the net effect of this assumption was to change the mean lake rainfall estimate by about 2–3% 

Close Modal

or Create an Account

Close Modal
Close Modal