Archive for the ‘Forecasting’ Category

California – The wettest start since 2010… And more on the way
January 31, 2017

I’m getting asked a lot of questions about our weather.  Why are we getting so much rain?  Why is it a year delayed from last year’s strong El Nino?  Is the drought over? What’s going on?

sofarthisyear

Four months into the water year we have the wettest start ever recorded.

To answer these questions I first have to give some perspective on climate in California and probabilities.

In case you don’t know, California is a state of extremes when it comes to precipitation.  Throughout recorded history we’ve alternated between drought and excessive rain.  Los Angeles is a good example of these extremes.  While we average 15 inches of rain a year, we are rarely near this amount.  If you look at the past 30 years, we’ve gone from extremely dry years to well above average years.  In the end, it averages 15 inches.

30yearsrain

snowpack

Next snow survey is February 2nd and we’ll near 200%.

The one question that is a little more difficult to explain is the why we’re getting the heavy rain this year instead of last year.  Last winter we were prepared for the potential of heavy rain. The reason we thought this was because in all of the significantly strong El Nino climate patterns, Southern California got soaked.  The odds were in our favor.  But the rain didn’t come.

2015_Significant El Ninos

The 2015/16 water year only recorded 6.57″ of rain.

What is fascinating this year is we are in a weak La Nina or neutral weather pattern.  Both of these patterns slant dry for the state of California and especially for Southern California.  The probability our state would be dry this year is 70%, leaving a 30% chance we’d be above average.  And, if you really study this image, there isn’t a weak La Nina pattern that is well above average for Southern California.

laninaneutralpattern

Instead what we are seeing is the wettest winter since 2010.  The storms of the past two months have recouped 37% of the state’s five-year snow/water deficits. And for the first time since January 2014, no place in California is in an exceptional drought.  Precipitation is more than 200% of average and the Sierra Snow pack is looking great!

droughtupdate2016_snowpack_update

So what gives? Why the two extremes in outcomes?  I could give the answer, “That is how the weather works sometimes.” And this reply wouldn’t be completely wrong because even with an excellent forecast for a 90% chance of rain, there are days it’s dry.

But the answer is a little more complicated.

Last year the water was so warm in the Pacific that it affected the jet stream and moved all of the heavy rain to the Pacific Northwest and Northern California.

This winter we’ve been aided by numerous atmospheric rivers or a Pineapple Express.  This type of pattern can occur at any time regardless of an El Nino, La Nina or neutral pattern.  These rivers of atmospheric moisture are responsible for much of the devastating flooding in California’s history.  And it played a big part in the flooding on Jan. 22.

longbeachflooding

Long Beach received the most rain ever recorded in a 24 hour period, 3.97″.

ar

Most people don’t know about the mega flood of 1862, which was caused by an intense atmospheric river. It has a return period of 100-200 years, meaning it will happen again.

califlooding

And this leads me to climate change.  The latest research shows that in a warming planet, droughts will become more severe and heavy rain events will occur more often.  Before we started getting the rain in December, Southern California went through the driest five-year stretch ever recorded.  And, if you look at tree ring data, it may have been the driest stretch in 1,000 years.

record-dry2016_recorddry

But if you look at the history of flooding in California, intense flooding events occurred before the industrial age and atmospheric rivers are responsible for most of these events.  Attributing climate change to future flooding events may be hard to do.  Upcoming research will need to address how a warming world affects atmospheric rivers.

Going forward, the key for our drought, especially in Southern California, is we need this kind of winter pattern for two more years so we can get out of a drought cycle.  And, of course, I’ll keep you posted.

 

Forecasting beyond 7-Days
May 24, 2013

Question-Mark

I get questions like these quite often:

“What are the predictions for June 6th through 10th for Houston/Dallas?  Is there a website for future predictions?” – Julia

“I was wondering what the weather will be like for next weekend before the Memorial Holiday? I’m taking a road trip to Abilene, Texas, and I’ll be driving through Dallas, and was wondering what the weather will be like when driving through Dallas.” – Rachel

“My husband and I are celebrating our 1st anniversary over Memorial Day weekend and have a trip to Cancun booked. I checked the weather today and looks like it is going to be storming EVERY day that we are there. I am kind of freaking out and trying to decide what to do.” – Leslie

The question is: How do I make plans for a forecast beyond seven days? There isn’t a comforting answer.

Weather models are pretty good three to five days out. However, there is a drastic drop in accuracy once you go beyond six days. All of these questions are asking for a forecast 9 to 20 days away, and a lot can change with the weather in that time.

Low pressure disturbances that create rain usually only show up on the weather models 12 hours to three days out. Cold fronts, especially for southeast Texas, can stall north of us or even on top of us, and that will completely change a forecast two to three days out. This explains why our seven-day forecast will change for the weekend Monday through Thursday of the same week. I could keep giving examples, but the point is: The weather is constantly in motion and always changing.

Studies have been done on long-range forecasting past seven days and found no skill in prediction. In fact, climatology, or the seasonal average, will beat a computer-generated long-range forecast.

AccuWeather is marketing a 25-day forecast. I would love a middle school school student to do their science fair project comparing the forecast to what actually happens.