Spring Forecast 2019

Forecast Overview

Temperatures

Temperatures for Spring 2019 will most likely be around 0.5C above the 1981-2010 long-term average over much of the country, with the north generally seeing larger positive anomalies than the south.

Forecast Factors

ENSO Signal

ENSO has been weakly positive during Autumn 2018 and during the winter of 2018/19, although despite this the atmospheric circulation during February 2019 has been more typical of La Nina conditions and the change to a cold easterly type, which had been widely forecast by the long-range outputs until the end of January, did not materialise during February, which instead ended up featuring frequent warm southerlies, and this in spite of additional help from a sudden stratospheric warming event.

ENSO is forecast to remain positive through the spring, but with some disagreement regarding how strongly positive, with the ECMWF ensembles suggesting a continuation of the +0.5-1.0 range, and the Met Office ensembles pointing towards an increase to +1.0-1.5. The NMME ensemble shows a large scatter towards the end of the season, emphasising the high level of uncertainty, but its ensemble mean broadly agrees with the ECMWF ensembles, suggesting values in the 0.5 to 1.0 range.

For previous years when a moderately positive winter ENSO (0.5 to 1.0) remained at that state in the spring or increased to the 1.0-1.5 range, notable analogues are 1980, 1991, 2004 and 2005. Spring 2015 started off with a similar ENSO state to this year, but developed into a strong El Nino by May, which currently looks unlikely to happen this year.

The Marches of 1980, 1991, 2004 and 2005 on average had an anomalous east-south-easterly flow over the British Isles with a high pressure anomaly over northern Scandinavia. This is not always a cold weather pattern, as it points to an above-average frequency of southerly winds which can bring rather mild weather as happened in March 1991, but the Marches of 1980 and 2004 did contain cold snowy "easterly" spells and the first half of March 2005 was cold, with snow for some. A more consistent theme was below-average sunshine in eastern areas. The Aprils of those years showed a very slight tendency for above average pressure and an enhanced westerly flow, but no clear signal. The Mays of those years were somewhat more anticyclonic than average with a high pressure anomaly centred in the eastern North Atlantic, giving an anomalous northerly component to the mean airflow over the British Isles. The Mays of 1980, 2004 and 2005 were all generally sunnier than average, but that of 1991 was notably dull in most central and eastern areas despite also being very dry. For spring as a whole, slightly above average mean sea level pressure is suggested, pointing towards a continuation of the below-average rainfall and concerns over water shortages as we head into the summer of 2019.

Stratospheric conditions

The QBO (Quasi-Biennial Oscillation) has moved into a weak westerly phase which may have contributed to the re-establishment of the polar vortex into February after the wintry spell of late-January. This may point towards an above-average chance of westerly types dominating early in the spring, but the effects of the QBO on the weather typically fade as the spring progresses.

MJO

The Madden-Julian Oscillation is currently forecast to head through phases 2 and 3 in the first half of March, possibly approaching phase 4 by midmonth. The composite for phase 2 gives an anomalous southerly flow over the British Isles, the composite for phase 3 is an anomalous easterly flow, and the composite for phase 4 gives an anomalous cyclonic north-westerly flow. The implication is that should the MJO influence become prominent towards mid-March we may see a period of weaker than average westerlies associated with phase 3, or possibly a temporary easterly incursion. Current model outputs are predominantly suggesting that the former is more likely with little evidence of pressure being likely to rise over Scandinavia. A shift into phase 5 later in March would promote the likelihood of another westerly spell, and if the MJO ends up in phases 5-6-7 in April, as is likely if the current rate of propagation continues, high pressure may be more prominent than usual. It is difficult to forecast where the MJO will end up by May.

Arctic Sea Ice

The Arctic sea ice extent is running higher than the previous three years, but like last year there is particularly low extent in the Chukchi/Bering Sea area. Sea ice extent in the Atlantic side of the Arctic is mostly near normal, and in January and February temperatures near the North Pole have generally been within a few degrees of the long-term average according to the ERA-40 reanalysis, in contrast to the large positive anomalies seen in 2016, 2017 and 2018. There are no clear-cut links between low Arctic sea ice extent and the weather of the following spring, but with the Atlantic side of the Arctic seeing close to average sea ice extent and temperatures near the North Pole running much closer to average than recent years, there is potential for any northerly outbreaks in early spring to be quite potent. Note for example that there was widespread snow from a predominantly north-westerly type in the closing days of January, a setup which could recur during the first half of March.

North Atlantic Oscillation and Arctic Oscillation

The North Atlantic Oscillation is generally forecast to be positive during the first half of March, while the Arctic Oscillation is forecast to start off positive but fall negative towards mid-March, which suggests a less organised polar vortex and perhaps increased potential for colder air masses to head towards the British Isles, but the continued positive NAO would imply quite a strong westerly influence continuing from the North Atlantic.

Temperature and rainfall forecasts

The Met Office probability and ensemble mean maps for March-May 2018 are pointing towards anticyclonic weather being more prominent than usual over the British Isles, and more especially so when the forecast period is moved forward to April/May/June, suggesting a continuation of the generally below-average rainfall that has persisted since June 2018. There is an implication that the most strongly anticyclonic conditions are forecast to be centred on May, which is also consistent with the ENSO analogues. The probability maps suggest relatively low confidence in the below-average precipitation for March/April/May, but confidence increases for the period April/May/June. Temperatures are forecast to be above the long-term average but with only small positive anomalies.

The Netweather long-range model is suggesting that March will be a little more cyclonic than average and with a slightly reduced mean westerly flow relative to normal. Near normal pressure is suggested for April and then a slight anomalous northerly component to the mean airflow for May with sea level pressure near to slightly below average. This is rather at odds with the forecast from the Met Office model, but the projection for March is consistent with the current sets of medium-range model outputs.

The CFS model suggests a NW-SE split for March and April with above average precipitation for the north-west but below average in the south, especially in April. A reversal of this is forecast for May with drier than average weather in the north-west and wetter than average weather for the south. Temperatures are forecast to be well above normal in March, moderately above in April and nearer normal in May, with the north seeing the largest positive temperature anomaly.

The ECMWF long range guidance has a weak signal for above average pressure especially to the west of the British Isles, and temperatures close to or just above normal for the period March to May 2019. As with the Met Office ensembles, there is a suggestion that the above average pressure could peak in May, and there is a very weak signal for below average precipitation for March-May.

The JAMSTEC model outlook has a colder than average spring over southern Britain with near average temperatures for the north, and below average rainfall across the whole country.

The CanSIPS model is going for a slightly reduced westerly flow relative to normal for March, consistent with the ENSO analogues and the Netweather model forecast, with cyclonic weather more dominant than normal in April and May. Precipitation is forecast to be near normal in March, and above normal in the south and near normal in the north in April and May (April wettest relative to normal in the south).

Conclusions

This is a tricky season to forecast due to the mixed signals, as well as the failure for easterly winds to materialise during February despite a large number of signals and forecast model outputs having suggested that February would be somewhat more "easterly" than usual. There is quite strong agreement on March having a reduced mean westerly flow relative to normal, with an anomalous east to south-easterly flow over the British Isles, but only weakly so, indicating that the predominant wind direction for March is still expected to be a westerly or south-westerly. Near average pressure, slightly above in the north and below in the south, is suggested, but it looks probable that the first half of March will be unsettled.

There are signals for above average sea-level pressure during April and May, with high pressure for May supported by the ECMWF, Met Office and the ENSO analogues, but possibly centred in the eastern North Atlantic, which would promote dry weather but not necessarily particularly warm, and probably sunnier relative to average in the west than in the east. The precipitation signals for March-May are rather mixed and weak, however, despite the tendency for high pressure to be mostly associated with below-average rainfall. There is a consistent suggestion from the outputs that temperatures across Britain will be near to slightly above average (positive anomaly of 0 to 0.5C).