According to climate science, Mozambique has experienced a more or less continuous warming to its climate of over 1.2°C over the last 130 years. The reality as evidenced by the actual temperature data is rather different. Like many places around the world, Mozambique has certainly seen some modest warming of about 0.6°C since 1980, but before 1980 the picture is uncertain, and on balance there may have been no significant warming at all in this period.
Mozambique has eleven long and medium weather stations most of which are located on, or near, the coast, as indicated in Fig. 78.1 below. This is same number of long and medium weather stations that was seen for Madagascar (see Post 77). However Madagascar also has another nine stations with over 300 months of data; Mozambique has only four.
Fig. 78.1: The (approximate) locations of the weather stations in Mozambique. Those stations with a high warming trend between 1901 and 2000 are marked in red while those with a cooling or stable trend are marked in blue. Those denoted with squares are stations with over 1000 months of data, while diamonds denote stations with over 480 months of data.
Of the eleven long and medium stations in Mozambique, two are long stations with over 1200 months of data, another two have over 1000 months of data, and the remaining seven are medium stations with over 480 months of data. Averaging the temperature anomalies from these eleven station records results in the mean temperature anomaly (MTA) for the region shown in Fig. 78.2 below. The method used to determine the MTA was the same as that used in previous posts and has been outlined in Post 47.
First the monthly reference temperatures (MRTs) were calculated for a suitable time interval, in this case the thirty year period from 1951 to 1980. This ensured that all eleven long and medium temperature records contained sufficient valid data in this interval for each month (a minimum of twelve years of data for each MRT is usually required). The MRTs for each set of station data were then subtracted from the raw monthly temperature readings to produce the anomalies for that station location. The anomalies from all eleven long and medium station records were then averaged to determine the MTA.
Fig. 78.2: The temperature trend for Mozambique relative to the 1951-1980 monthly averages based on an average of anomalies from stations with over 480 months of data. The best fit is applied to the monthly mean data from 1913 to 2012 and has a positive gradient of +1.07 ± 0.06 °C per century.
The MTA data in Fig. 78.2 appears to show an upward trend of about 1°C per century, but that is not the whole story. As Fig. 78.3 below indicates, the number of stations that contribute to the MTA drops significantly as you go back in time before 1950. This in turn suggests that the trend in Fig. 78.2 after 1960 is much more reliable than that before 1930.
Fig. 78.3: The number of station records included each month in the mean temperature anomaly (MTA) trend for Mozambique in Fig. 78.2.
As usual I have compared my results based on raw temperature data with the Berkeley Earth (BE) version based on their adjusted data. The equivalent average of the BE adjusted anomalies is shown below in Fig. 78.4. From 1940 onwards it is very similar to the trend from the raw temperature data shown in Fig. 78.2 above. In both cases the MTA plateaus between 1940 and 1980 before rising by about 0.8°C over the next twenty years. Then it drops back slightly by about 0.2°C in the following decade.
Fig. 78.4: Temperature trends for Mozambique based on Berkeley Earth adjusted data. The average is for anomalies from all stations with over 360 months of data. The best fit linear trend line (in red) is for the period 1911-2010 and has a gradient of +1.00 ± 0.03°C/century.
The MTA based on BE adjusted data also shows good agreement between 1920 and 1990 with the official Berkeley Earth trend for Mozambique shown in Fig. 78.5 below. It is only after 1995 that there is significant disagreement between the two plots. This may be because Berkeley Earth has included more data in the average after 2000 using many more stations with less than 300 months of data than I have.
Fig. 78.5: The temperature trend for Mozambique since 1840 according to Berkeley Earth.
It is clear that the temperature trend for Mozambique exhibits some warming after 1980, but the trend before 1980 is less clear. However, if we look at all the available data (see here for a list of all stations in Mozambique) we do see that there are a number of stations with data from 1930 to 1960 that are not included in the MTA trend shown in Fig. 78.2 above, due to insufficient data within the MRT period. It turns out that the temperature time series of most of these station datasets display cooling trends from 1930-1960. The one main exception is the data from Lourenço Marques (Berkeley Earth ID: 156935) which shows more or less continuous warming from 1920 onwards.
If we wish to incorporate these extra stations into the MTA then we need to change the MRT period to the interval 1931-1960 when most of this extra data was recorded. This is what I have done to generate the data in Fig. 78.6 below. The other significant change was to discard the data from Lourenço Marques (aka Maputo) from the MTA. The reason for doing this was that, while this station clearly has the longest set of temperature data in Mozambique (it has temperature data from as far back as 1892), its temperature trend is an outlier. This is probably because the station is located in the capital city Maputo and is more prone to the urban heat island (UHI) effect than most other stations in the country. I should note that this outlier behaviour of capital cities is not unusual. The same was seen for Jakarta in Indonesia (see Post 31) as well as for Sydney and Melbourne in Australia.
Fig. 78.6: The mean temperature trend relative to the 1931-1960 monthly averages for stations in Mozambique with over 300 months of data but excluding Lourenço Marques. The best fit is applied to the monthly mean data from 1921 to 1980 and has a positive gradient of +0.16 ± 0.12 °C per century.
The result of this change of MRT, together with the removal of the data from Lourenço Marques, is to completely change the temperature trend before 1940 (see Fig. 78.6 above). While the trend after 1980 still exhibits a warming of 0.6°C, the trend before 1980 is now neutral but with significant natural variability. It is also more reliable as it is based on data from more stations than previously (see Fig. 78.7 below). This would suggest that the temperature trend in Fig. 78.7 is a better representation of the true behaviour of the Mozambique climate over the last 100 years than is the trend in Fig. 78.2. It also provides more evidence to suggest that anthropogenic climate change is really only a post-1980 phenomenon and not one that began in 1850.
Fig. 78.7: The number of station records included each month in the mean temperature anomaly (MTA) trend for Mozambique in Fig. 78.6.
Summary
The data shows that there is clearly strong warming of almost 0.6°C in Mozambique after 1980 (see Fig. 78.2).
Before 1980 the climate appears to be stable (see Fig. 78.6).
Overall this suggests that the warming seen in Mozambique after 1980 is probably real as it coincides with the period of greatest anthropogenic carbon dioxide production. It is also similar in timing and magnitude to temperature rises seen elsewhere around the world as I have shown previously.
Acronyms
BE = Berkeley Earth.
MRT = monthly reference temperature (see Post 47).
MTA = mean temperature anomaly.