Digging Down Deeper into the Data

 In the previous post, I used this graphic (Figure 1) from the UK Met Office showing the mean temperature anomaly for December 2023. For England, the mean temperature anomaly was predominantly (dark pink shading) between +1.5 ℃ and + 2.5 ℃. There were, however, some hot spots (e.g. the Thames Valley and Cotswolds circled in black) where the mean temperature anomaly (red shading) was between +2.5 ℃ and + 3.5 ℃. 

Figure 1: Mean Temperature Anomaly for December 2023 (Met Office)

Those of a curious nature, or indeed just genuinely sceptical, may wonder/question if this was correct. Fortunately, it is very easy to track down the data and prove it to yourself that it is, in all probability, true.

The first step is find the data - an easy task for the UK as the Met Office makes it readily available. Here are the data for Ross-on-Wye and Oxford weather stations that fall within, or close to, this Thames Valley/Cotswolds hotspot.

The second step is to put the data for the two sites into a spreadsheet (I used Google Sheets) and calculate the 1991 - 2020 baseline temperatures for December by averaging the mean temperatures for this particular month. For Ross-on-Wye and Oxford, the average December temperatures over the 1991 - 2020 period were 5.7 ℃ and 5.4 ℃ respectively. It is then a simple procedure to calculate the temperature anomalies by subtracting these average temperatures from the observed December temperatures at each location. The results are plotted out below:

Figure 2: Temperature Anomaly Plot for Ross-on-Wye Weather Station

Figure 3: Temperature Anomaly Plot for Oxford Weather Station

What can we deduce from these plots?

1. December 2023 was indeed very hot compared with the 1991 - 2020 average. The anomaly for Ross-on-Wye was 2. 8 ℃ and that for Oxford was 2.6 ℃.
2. The anomaly profiles of the two weather stations are similar. This is to be expected since the two stations are less than 100 km apart.
3. Mean monthly temperatures for December fluctuate; sometimes colder and sometimes hotter than the 1991 - 2020 average
4. However, for the past decade or so (2011 to date), the frequency of 'colder than normal' Decembers has decreased noticeably.
5. Mean December temperatures at both locations are trending upwards at a rate of about 0.5 ℃ per decade (low confidence)

In the early period (1991 to 2011) there appears to be a cyclic pattern moving from negative to positive and back again (Figure 2 and Figure 3). This cyclic pattern is less discernible after 2011.

Cyclic patterns or oscillations in atmospheric and ocean currents are well-known; e.g. the El Nino Southern Oscillation (ENSO). The observation that 2023 is the hottest year globally in the modern record has been attributed, in part, to ENSO being in its El Nino phase. ENSO has its greatest influence on the Pacific region with any impacts on the UK/Europe being very small.

On the other hand, the North Atlantic Oscillation (NAO), as the name suggests, impacts the continents & countries surrounding the North Atlantic. When the NAO is in its positive phase, it brings warm, wet weather from the west to the UK. In its negative phase, the NAO brings colder drier winds from the east and north-east.

So lets see if there is a correlation between the December temperatures in the UK and the phase of the NAO. I will reuse the mean December temperature data from the two weather stations: Ross-on-Wye and Oxford. There are a number of websites where a record of the NAO phases can be found. I have used this NOAA site from which to download the data (see Figure 4 for a snapshot).

Figure 4: Monthly NAO Indices for 1950 to 1964

After filtering out the December NAO indices (far right of Figure 4), I added this information to my spreadsheet containing the December temperature data for the two weather stations: Ross-on-Wye and Oxford. The scatter plots (mean December anomaly vs NAO Index) with least squares regression trendlines are shown below in Figures 5 & 6.

Figure 5: Correlation of December Anomalies (Ross-on-Wye) with December NAO Index

Figure 6: Correlation of December Anomalies (Oxford) with December NAO Index

I was surprised at how good the correlation between December mean monthly temperature anomalies and the NAO indices was. We can conclude the NAO index is one of a number of parameters that influence the average monthly temperatures in December. The NAO index describes changes in atmospheric pressure patterns between two regions; one near Iceland and one near the Azores. Given plenty of computing power, short-term weather predictions can be made based on the pressure fluctuations. Unfortunately, long-term predictions (e.g. will December 2024 be warm or cold) are not possible.