NOAA Provides Clarification On The February, 2020 Temperature Forecast

Written by Sig Silber

At the end of every month NOAA issues an update to their Early Outlook for the following month. The Early Outlook for February 2020 was issued as usual on the third Thursday of the preceding month (January). After reading the full month and the Week 3 – 4 discussion, in our view there remained some significant uncertainties, especially in terms of the temperature forecast for California and Nevada. We reached out to NOAA and they have provided some additional information that we are presenting in this article.

Please share this article – Go to the very top of the page, right-hand side for social media buttons. Also, feel free to send this article to anyone you feel will benefit from it. You can find the latest version of all our weather articles by consulting the Directory by clicking here and then clicking on the latest version of the article which is of interest to you.

First, we will start with a summary of the Early and Updated forecast and then we will provide the additional information provided to us by NOAA.

Here we compare the prior forecast to the new forecast as above but in a more compact format. In the original article, we showed the four partial forecasts but those were set up to auto-update so it would take a lot of effort to reconstruct what was published last Friday night. We did not conclude that this was necessary.

And with respect to drought, this was also issued on January 31, 2019.

Below are the comments from NOAA which were contained in three emails. I have deleted the names and contact information of those copied to avoid them receiving a lot of email whenever anyone has a question on a forecast. I put what I considered to be the most important comment from the key forecaster in bold type.

Additional Information Provided by NOAA

Hello,

I was the forecaster for CPC’s updated February outlook, released last Friday, and was forwarded your comments below. (See text copied in red). You raise some interesting questions regarding how the various CPC outlooks could be consistent. I will try to address your comments regarding the temperature forecasts for California and Nevada.

Forecasts for the first week of February indicated above normal temperatures were likely. Although the CPC 6-10 day temperature outlook indicated that below normal temperatures were more likely than above normal temperatures when averaged over the period, the probabilities were relatively modest for this region in the 6-10 day outlook, while increasing in the week 2 outlook. This transition was described in the February forecast discussion and related to potential impacts of the predicted state of the AO and MJO. It should be noted that while the AO and MJO model forecasts were considered in the week 3-4 outlook, the outlook indicated equal chances of above and below normal temperatures over California and most of Nevada, and only a very weak tilt towards below normal temperatures from northern Nevada northward. Also, the impacts of the MJO would be expected to vary from week 2 through weeks 3 and 4, and the state of the MJO and AO are more uncertain later in February.

Because the week 3-4 outlook is equal chances, it does not alter the chances of above or below normal for the month as a whole. The forecast for the month of February over California and Nevada is largely determined by the first half of the month, with the greatest skill coming from the first week of the month. It should be noted that the forecasted probabilities in the outlooks cannot be translated into average anomalies or a full probability distribution of possible anomalies, without consideration of the skill at each lead time. CPC does consider the skill of forecasts for various lead times when updating the monthly outlook. This is done both subjectively, by the forecaster for the official outlooks, and empirically by a consolidation of model forecasts at various leads.

Likely above normal temperatures for February in parts of the west are consistent with the greater skill of week 1 forecasts compared to week 2 forecasts, and the low skill and uncertainty in the week 3-4 outlook. The February temperature outlook for this region is also consistent with the empirical consolidation of model forecasts at various lead times, which uses the skill at the various lead times to determine weights and probabilities. Lastly, the February outlook for this region is consistent with slower varying climate signals from seasonal to decadal time scales that impact CPC seasonal outlooks. Slower varying climate signals have a greater influence on the probabilities of above and below normal terciles for the entire month than average temperatures for weeks within the month.

For the February drought outlook, you might like to refer to the forecast discussion here:

https://www.cpc.ncep.noaa.gov/products/expert_assessment/mdo_discussion.php

The monthly drought outlook discussion notes consideration of all CPC temperature and precipitation outlooks, and concludes that: “drought development is possible in California”, especially in currently abnormally dry regions, “but more likely to occur in March or April.” The discussion notes that snowpack is “above normal in much of the West” and “major California reservoirs are in good shape”, suggesting that it may take more than a dry and mild February to develop drought conditions, since “…March and April remain in the wet season…” for this region.

Please contact me if you need further information.

Best regards, Dan Collins

Dan C Collins, PhD

NOAA Climate Prediction Center

NCWCP 3021

5830 University Research Court

College Park, MD 20740

And then a followup communication

Sig,

That is fine to include my response in an update of your article. You are probably aware that I was not the forecaster for the monthly drought outlook, so my comments on that forecast are drawn from the released forecast discussion as well as interaction with the drought forecaster, as with the other outlooks and forecasters. You may also want to consider that the next seasonal outlook will be issued on the 13th of this month, and you may wish to update with the latest information.

About your last email, it is correct that we do not have temperature probability forecasts in products for days 1-5. As I mentioned, we do have some model-based forecast tools for the monthly update that consider the periods within the month separately. Also, the relationship of current temperatures to the 1981-2010 climatology is a greater component of the forecast in some areas and seasons than other areas. We have statistical tools to examine this. This has resulted in a greater chance of above normal in California and Nevada at this time of year. The 1991-2020 climatology will change the temperature threshold considered above normal, and this will mean a reduction in the probabilities of above normal for many areas, including California and Nevada in this season.

Best, Dan

And one more

Sig,

Quickly wanted to clarify. I wrote the monthly February temperature and precipitation update. {Name deleted} wrote the half month lead, earlier February outlook. {Name deleted} wrote the Monthly Drought Outlook discussion.

Best, Dan

I did not include my replies as it was not my intent here to have a debate (and Dan Collins pretty much included what was in my emails in his responses) but simply to provide some additional information from one of the key forecasters. I was not convinced but I found the comments helpful so I thought I would pass them along. We all know that weather is a very dynamic situation and the forecast is subject to continually being updated. We could do a post mortem at the end of February but I think there will be new issues by then that are more important than the question I raised about the California and Nevada temperature forecast for February.

Conclusion

The purpose of the original article was to present the updated Outlook for February 2020 and compare it with the Early Outlook. It was not to critique the updated Outlook for February based on our opinions but we point out possible inconsistencies when we find them. We found a few including, of most importance, the one discussed in this article. Often (but not always) when I see what looks like an inconsistency, I ask for assistance from NOAA to inform me if I had noticed something of interest or just misinterpreted what I read.

At this point, I feel that I had noticed something of interest but I now better understand why at least for California the forecast was made the way it was made. It points out the great difficulty forecasting presents and I hope this sheds some light on the process.

Many thanks to Dan Collins for taking the time to respond to my inquiry.

.