Seasonal Forecasts Published April 18, 2020: Part I, NOAA Four-Season Forecast

Written by Sig Silber

Here is the April 16, 2020, NOAA 15-month forecast for CONUS and Alaska including the Early Outlook for May 2020 and forecasts out through July 2021. The initial part of the forecast is very similar to what was issued last month. NOAA is not very confident in their Early Outlook for May, especially precipitation, so we might expect a lot of revisions when that is updated on April 30. Looking beyond May, the temperature forecasts are not much changed but the precipitation forecasts are very different.

 This article was put together very quickly so I have not had as much time to study the NOAA discussion but I am pretty sure what has taken place is increased confidence in ENSO becoming more like a traditional weak La Nina or ENSO Neutral with a strong La Nina bias. Soon (possibly Sunday Night) we will compare the NOAA forecast to the JAMSTEC forecast. We will discuss the ENSO situation more at that time but right now we tend to agree with NOAA and expect that JAMSTEC does also but their ideas of the impacts may be somewhat different. We will see.

Please share this article – Go to the very top of the page, right-hand side for social media buttons. Please feel free to share this report with anyone you feel will find it useful.

A. Focus on the NOAA Update

A note about terminology; the deviations from climatology/normal are color-coded but also labeled “A” for more than (above) normal and “B” for less than (below) normal. The area designated EC means Equal Chances of being more or less than normal. In my comments For the shorter-term maps, NOAA uses “N” instead of “EC” as they believe they can be more definitive with respect to shorter timeframes. So the words “warm”, “cool”, “wet”, “dry” should be generally interpreted as being relative to climatology/normal for that location and time of year.

First, we will take a look at the NOAA Early Outlook for May,  2020. It is called the Early Outlook because it will be updated at the end of April. Only the May Outlook will be updated at that time.

Temperature

Precipitation

We have nothing to compare these maps with as NOAA does not provide in their Update a forecast for each of the subsequent two months separate from their three-month forecast. Thus I do not have forecast maps for May from the previous NOAA Report to compare against. And the current month is not over so we can not really compare the forecast for next month against the actual for this month. It is probably best to just try to understand what NOAA is trying to convey about May*. For Temperature, it is mostly warm for Alaska and the Alaskan Panhandle and for most of CONUS except for a large Central Area that is EC. With respect to precipitation, southeast Alaska is dry as are the western parts of Washington and Oregon.  The Southeast is wet as is most of the Great Plains and some of the northern Rocky Mountain states. 

* compared to normal/climatology for the indicated time of the year.

Visual Consistency Testing.

It is useful to see how the month forecast fits with the set of shorter forecasts that we have. The first map covers mostly April so only two of the three maps are really useful for this purpose and the second map  barely impacts May. The third map only extends the coverage through the first 15 days of May. The Seasonal Outlook was issued fairly early this month since the third Thursday fell on April 16, 2020. I think the latest it can be is the 21st of a month. Fifteen days of shorter forecasts is not a solid basis for doing this visual consistency testing but it is what we have to work with. It may tell us how the last sixteen days of the month should differ from the first fifteen days for the full-month forecast to be accurate.

First Temperature

The new monthly forecast is to the right. Does it appear to be reflective of the parts of the month for which we have forecasts? →		↑ ← The full month forecast fits only fairly well with the Week 3 – 4 forecast which by itself only covers fourteen days of May as of when we published this article. Will the Great Lakes cool anomaly be neutralized in the second half of May? We will revisit this question and others at the end of April when the May forecast is updated. We should also have a better feel for the situation next Friday.

And then Precipitation

The new monthly forecast is to the right. Does it appear to be reflective of the parts of the month for which we have forecasts? →		↑ ← The full month forecast fits only fairly well with the Week 3 – 4 forecast which by itself only covers fourteen days of May as of when we published this article. Will the Northern Tier dryness moderate in the second half of the month. We will revisit this question and others at the end of April when the May forecast is updated. We should also have a better feel for the situation next Friday.

Now we consider the three-month Outlook.

Notice that the three-month periods are abbreviated e.g. May/June/July is shown as MJJ You will see such abbreviations often in this report.

Prior Temperature Outlook for MJJ 2020

New Temperature Outlook for MJJ 2020

It might look cleaner with smaller graphics. Here goes.

Now Precipitation. 

Prior forecast for MJJ 2020

New Precipitation Outlook for MJJ 2020

It might look cleaner with smaller graphics. Here goes.

Now let us focus on the long-term situation.

First Temperature

Prior 14 Month Temperature Outlook: MJJ 2020 – AMJ 2021

New 14 Month Temperature Outlook: JJA 2020 – MJJ 2021

To compare maps from one release to another, one needs to remember that the new release drops one three-month period and adds a later one. So to make the comparisons one has to shift the new maps to the right one position and that makes the map on the right drop down to become the left-most map in the next level. I do not have a computer software tool for doing that for you so you have to do it mentally. When I do the comparison, I print the two sets of maps and put them side by side and number the same three-month maps 1, 2, 3,…..,11 in both sets of maps to make it easier for me to easily compare the same three-month period in the new with the previous forecast. One uses the same procedure to compare the precipitation maps. Based on this procedure, I conclude that:

Now Precipitation

Prior 14 Month Precipitation Outlook: MJJ 2020 – AMJ 2021

New 14 Month Precipitation Outlook: JJA 2020 – MJJ 2021

If you want larger versions of each map (temperature and precipitation) you can find them here. And then each of those maps can be clicked on to further enlarge them.

Sometimes it is useful to compare the three-month outlook to the forecast for the first of the three months. It shows how much the pattern changes over the three-month period.

One can mentally subtract the First-Month Outlook from the Three-month Outlook and create the Outlook for the last two months in the three-month period.

*The concept is that the probabilities of a deviation from climatology in the First Month and the combined Month Two and Three forecast that one derives must average out to the probabilities shown in the three-month maps.

Below is the NOAA Discussion (slightly reorganized) released by NOAA on April 16, 2020. Headings that are “Initial Cap” only rather than all caps were added by the Author of this Update Report for clarity. Also, we have organized the sequence of the sections of NOAA Discussion to first present the Atmospheric and Oceanic Conditions and then the Initial Month (May), the new NOAA Summary for MJJ, and finally the remainder of the 15-Month Forecast.

CURRENT ATMOSPHERIC AND OCEANIC CONDITIONS

ENSO-neutral conditions are in place across the Pacific Ocean at the current time. Sea surface temperature (SST) over the equatorial Pacific are almost entirely above normal, with a large region of anomalies exceeding +1.0 C centered on the Date Line. The Nino3.4 regional SST anomaly continues to hover around +0.5 C. At depth, positive ocean temperature anomalies dominate between 0-100 meters, except over the far eastern Pacific. Negative temperature anomalies are now observed between 100-200 meters depth from 150E to about 120W. Total 0-300m upper ocean heat content anomalies have turned negative for the first time since September 2019.

Atmospheric conditions now show suppressed convection over the past month near and east of the Date Line. This convective anomaly is consistent with symmetric off-equatorial 200-hPa cyclonic circulations.  Near equatorial zonal wind anomalies show some strengthening of the Walker Circulation over the central and eastern Pacific.

 Anomalously high soil moisture content is present from the northern Plains to the western Great Lakes southward to include most of the Great Plains, Mississippi Valley, Ohio Valley, Tennessee Valley and Southeast. Anomalously dry soils are currently observed over the Florida Peninsula and entire Gulf Coast including South Texas, as well as over much of the Pacific Northwest.

Coastal SST anomalies are small in magnitude this year for waters in proximity to the Alaska south and west coasts and sea ice coverage in the Bering Sea and Bering Strait are for the first time in a while generally near the long term average – although ice is primarily first year ice, is thin and likely to melt out relatively quickly. Coastal SST anomalies near the West Coast of the CONUS are likewise low amplitude. By contrast, the Gulf of Mexico is much warmer than normal, as are waters off most of the Eastern Seaboard.  

PROGNOSTIC DISCUSSION OF SST FORECASTS

The official CPC Nino3.4 consolidation forecasts a shift toward weak La Nina conditions by late summer and fall, due largely to the CFSv2 which continues to forecast developing La Nina conditions this summer. The SST constructed analog forecast largely follows the consensus forecast and nicely smooths over the jump between OND and NDJ where the CFSv2 times out of the forecast consolidation. The other two statistical tools (Markov and CCA) continue with warm neutral or weak El Nino during the entire forecast period. The NMME forecast overall follows the consolidation through SON, though it is interesting to note that three constituent modeling systems clearly favor La Nina (including the CFSv2), three favor neutral, and one is in between. The CPC-IRI consensus forecast is close to the objective consolidation, but slightly more conservative. It nonetheless favors a clear transition from  warm-neutral to cold-neutral ENSO conditions over the next several months.   

It is interesting to note that the latest experimental tool we have is an artificial neural network model that quite strongly favors the La Nina outcome in this uncertain situation.

30-DAY OUTLOOK DISCUSSION FOR MAY 2020

The May 2020 temperature and precipitation outlooks are based primarily on dynamical model guidance, statistical tools, and current soil moisture conditions. ENSO is not a factor in the monthly outlooks since it is predicted to remain neutral. Since the beginning of April, a large-scale area of enhanced convection propagated rapidly eastward across the global tropics from the Maritime Continent to the Western Hemisphere. As of April 15, the anomalous upper-level divergence and associated enhanced convection are overspreading Africa. Although a coherent Wave-1 pattern of enhanced (suppressed) convection is observed over the Western Hemisphere (Indian Ocean and Maritime Continent), this anomalous convection is likely more related to a convectively coupled Kelvin wave due to its fast eastward propagation. Based on the latest dynamical model forecasts of the RMM index, Kelvin and equatorial Rossby waves are expected to contribute to anomalous tropical convection heading into May. Therefore, the Madden-Julian Oscillation (MJO) is unlikely to influence the mid-latitude circulation pattern. Weekly sea surface temperature (SST) anomalies, centered on April 8 are running as much as 2.5 degrees C above normal in parts of the Gulf of Mexico. Since SSTs (as of April 13) are near or slightly above 27 degrees C in the southern Gulf of Mexico, the future progression of Kelvin waves will have to be closely monitored for an early season tropical cyclone later in May into the beginning of June.  

Temperature

During the final week of April, the GFS, Canadian, and ECMWF ensemble means feature a less amplified longwave pattern compared to early to mid-April. Also, there is large spread among ensemble members which is not surprising due to the lower amplitude. The increased chances of above normal temperatures forecast across the western CONUS, Gulf Coast, and most of the East Coast are based on the North American Multi-Model Ensemble (NMME) and is also consistent with long-term trends  during the spring season. The largest probabilities (60 percent or higher) of above normal temperatures are forecast across central and south Florida due to unanimous support from the dynamical models , long-term trends, and much above normal SST anomalies surrounding the Florida Peninsula. Although the NMME indicates increased chances of above normal temperatures across the Great Lakes, equal chances (EC) of below, near, or above normal temperatures are forecast for this region since the ECMWF model features below normal temperatures during early May. Soil moisture currently ranks above the 90th percentile from the Missouri River Basin south to the Ozarks and Tennessee Valley. This high soil moisture content is likely to have a cooling effect on temperatures for these areas during May. The EC forecast for temperatures across the Great Plains, Mississippi Valley, and Corn Belt is consistent with these ongoing soil moisture conditions along with the calibrated NMME.

Precipitation

The dynamical models  are in good agreement and feature increased chances of above normal precipitation across the central Rockies, much of the Great Plains, and Southeast. This wet pattern is also supported by the CFS and ECMWF models (initialized on Apr 13) with their depiction of an upper-level trough over the Rockies and High Plains during early May. Despite the good model agreement, the probabilities of above normal precipitation are tempered due to predictability in the monthly precipitation outlook at this time lead. Also, precipitation becomes more convective and less predictable later in the spring. The NMME supports increased chances of below normal precipitation across the coastal Pacific Northwest. Modifications to the precipitation outlook are likely with the updated release on April 30.

Above normal temperatures are most likely across Alaska, due to good agreement among the dynamical model guidance. In addition, the CFS and ECMWF models feature above normal temperatures during early May. Probabilities of above normal temperatures were reduced across northern mainland Alaska, due to a deep snowpack for this time of year. The only notable signal among the precipitation tools is a slight lean towards below normal precipitation across the Aleutians and Alaska Peninsula during May.

SUMMARY OF THE OUTLOOK FOR NON-TECHNICAL USERS (Focus on MJJ)

ENSO-neutral conditions remain in place across the Pacific Ocean and ENSO-neutral conditions are favored to continue through Autumn 2020. However, there is some tendency toward La Nina conditions developing during autumn and early winter when the odds of La Nina (30-40%) exceed both that of El Nino and the climatological odds of La Nina. The ENSO forecast is quite uncertain from summer through the upcoming winter.  

Temperature

The May-June-July (MJJ) temperature outlook favors seasonal mean temperatures to be above normal for nearly all of the United States with the greatest odds forecast for southwestern Alaska, much of the interior West, along the Gulf Coast and for the eastern one-third of the country. Lower probabilities for above-normal temperatures are forecast for areas of the central lower 48. A similar pattern is forecast through the summer and early autumn months moving forward. By late autumn through winter, above-normal temperatures are favored across the southern tier of the contiguous U.S. with weaker temperature signals  forecast across the northern tier. This pattern is due to a combination of long-term trends  and a forecast shift toward colder SSTs over the central and eastern tropical Pacific.

Precipitation

For precipitation, there are elevated odds for above-normal seasonal total precipitation amounts for northwest Alaska and for most of the central and eastern U.S. Below-normal precipitation is most likely for the Alaska Panhandle and points southward and eastward to include the Pacific Northwest, northern California, and the northern Great Basin. Progressing into and through summer 2020, the aforementioned regions of above- and below-normal precipitation are largely maintained, but the east-west gradient take on a north-south structure during autumn through early winter.

 BASIS AND SUMMARY OF THE CURRENT LONG-LEAD OUTLOOKS

PROGNOSTIC TOOLS USED FOR U.S. TEMPERATURE AND PRECIPITATION OUTLOOKS

The temperature and precipitation outlooks for MJJ 2020 were based on a few factors, primarily bias-corrected and calibrated dynamical model guidance, statistical forecast tools utilizing various methodologies (including “bridging techniques”) and recent trends . Objective, skill-based combinations of the above information was weighted quite heavily in the current set of outlooks. ENSO-neutral conditions are still generally assumed, but given the transition toward negative SST anomalies over the equatorial Pacific that is forecast, there have been subtle changes to reflect the fact that El Nino is not likely from summer through winter (only about a 20% chance). Recent climate trends  are considered for all leads but were relied on more heavily for the middle and later forecast leads. Simultaneous and lagged relationships between soil moisture anomalies and subsequent temperature and precipitation anomalies are likewise considered.

For the first several leads, diagnostic tools based on the NMME are utilized to better understand the source of forecast climate anomalies. These tools use the historical skill of predicting the large-scale 200-hPa circulation patterns observed during the various seasons to generate a type of post-processed forecast the is reconstructed based on these leading patterns of variability. This analysis indicates that large-scale circulation patterns more consistent with cold ENSO conditions are forecast through SON, the last full season in the analysis.

PROGNOSTIC DISCUSSION OF OUTLOOKS – MJJ 2020 TO MJJ 2021

TEMPERATURE

The MJJ 2020 temperature outlook favors seasonal mean temperatures to be above-normal for nearly all of the United States. A region in the north-central CONUS is designated as “Equal Chances” (EC) where forecast probabilities for either above-, near- or below-normal temperatures are no different than the 33.3% climatological odds for a 3-class system. The forecast is consistent with long-term trends , but is actually cooler than the trend over parts of the central and southeastern CONUS. This is due in part to the expected effect of anomalously wet soils on the local energy balance and is not supported by the forecast upper-level circulation. The greatest odds for above-normal temperatures is for southwestern Alaska, the interior West, along the Gulf Coast and for the eastern one-third of the country. Overwhelming agreement between numerous seasonal forecast dynamical models  across a number of operational forecast centers and general consistency from previous months’ forecasts strongly support this outlook. Also, nearly all statistical guidance is consistent with the dynamical model guidance, with the SST constructed analog depicting the coolest solution over the central CONUS. Strong long term positive trends  (especially in the eastern CONUS, parts of the West and Alaska) contributed to the elevated odds of above-normal temperatures.

A similar forecast pattern persists through JAS, with the weakness over the central CONUS shifting southeastward based on the latest guidance from both dynamical and statistical tools. By ASO and SON, long-term trends  along with some weak tilt toward La Nina (and tilt away from El Nino), favor increased probabilities of above-normal temperatures across much of the interior West extending eastward across the central CONUS. Relative to the outlooks issued last month, there is very little change through the winter into next spring except to increase probabilities favoring above-normal temperatures over parts of the southern tier of the CONUS. Likewise probabilities favoring above-normal temperatures were decreased across parts of the Northwest and southern Alaska. Again, this is a modest reflection of the forecast tilt toward negative SST anomalies over the Nino3.4 region.

PRECIPITATION

The MJJ precipitation outlook depicts elevated odds for above-normal seasonal total precipitation amounts for northwest Alaska and for most of the central and eastern CONUS. The primary driver for the outlook across the central and eastern U.S. is the objective consolidation of the statistical tools and NMME, and is consistent with the long-term trend. The NMME forecast is a bit wetter over the northern and central Rockies and looks very similar to the lagged response to El Nino over that region. Analysis of the NMME circulation fields and associated reconstruction favors a much drier solution over the western and central CONUS. This increased confidence in using the consolidation forecast solution.

Below-normal precipitation is most likely for the Alaska Panhandle southward   and eastward to include the Pacific Northwest and northern California. Similar information as noted above is utilized as the basis for the outlook in these regions. Below-normal soil moisture content to a small degree increased the odds for below-normal precipitation for parts of California and Oregon due to potentially less local recycling of boundary layer moisture.

Progressing through the summer and autumn seasons, the precipitation forecast takes on a more north-south orientation with above-normal precipitation favored over the northern tier of the U.S. with below-normal precipitation favored farther south. Unlike the MJJ and JJA forecast, by SON the NMME precipitation forecasts are largely consistent with the forecast reconstructed by the forecast circulation patterns. This is consistent with the extratropical circulation taking on a more La Nina flavor at times and leads to increased coverage for the autumn and early winter months relative to previous outlooks. Long-term trends  favor modest probabilities of below-normal precipitation over parts of Southwest during late winter through spring. Forecasts for FMA through MJJ 2021 reflect the consolidated guidance which is largely informed by long-term trends .

New Seasonal Drought Forecast

ENSO Considerations

Here is the NOAA Proprietary ENSO Model which they generally do not use but which they seem to favor this month.

The below is probably the probabilistic forecast NOAA used.

Let’s take a look at the subsurface along the Equator. First I show the graphics for Fevruary and March

February 17, 2020	March 14, 2020

Here is the new version

Let’s look at the Australian BOM Nino 3.4 Forecast

And the JAMSTEC evaluation

Here is the short JAMSTEC discussion. We will discuss it a bit more on Sunday. For our purposes tonight I have only highlighted the comments that apply to Alaska and CONUS.

ENSO forecast:

Observation shows a mixture of the El Niño Modoki-type and the canonical-type at present. The SINTEX-F predicts that the tropical Pacific will return to a neutral-state from summer.

Indian Ocean forecast:

Observation shows that the tropical Indian Ocean is warmer-than-normal at present. The model predicts that the present state will persist through the latter half of this year. However, there is a large uncertainty in the prediction (30% of the ensemble members actually predict a positive IOD event).

The strong negative Indian Ocean Subtropical Dipole still persists. The model predicts that it will disappear from summer.

Regional forecast:

On a seasonal scale, the SINTEX-F predicts that most part of the globe will experience a warmer-than-normal condition in boreal summer and autumn.

As regards to the seasonally averaged rainfall in boreal summer, a drier-than-normal condition is predicted for most part of U.S.A., northern part of the South American Continent, South Africa, southern part of West Africa, Indonesia, and southwestern China. In contrast, Mexico, most part of Southeast Asia, Philippines, India, and Sri Lanka will experience a wetter-than-normal condition. In boreal autumn, a wetter-than-normal condition is predicted for most part of Southeast Asia, India, Philippines, and Indonesia. In contrast, central U.S.A., Mexico, most part of the South American Continent, southern Africa, southern part of West Africa, and some parts of China will experience a drier-than-normal condition.

The model predicts most part of Japan will experience warmer and wetter-than-normal condition in summer as a seasonal average. In autumn, most part of Japan will be warmer and slightly wetter-than-normal.

B. Conclusion

There were some fairly minor changes to the three-month forecast and we did not have a prior one-month forecast to compare to but  May, June and July are forecast to be fairly similar.

The rest of the forecast is quite different especially with regards to precipitation as the theme now is solidly a weak La Nina or ENSO Neutral with a strong La Nina Bias.