Wednesday, August 20, 2014

Advancing Night

The long summer day is now fading in Arctic Alaska, as a daily period of darkness advances north across Alaska with the change of seasons.  Last night was the first night that civil twilight was observed at Barrow, i.e. the sun briefly dipped more than 6° below the northern horizon.  However, it will be another 2-3 weeks before it starts to get reasonably "dark" at night in Barrow.

Here are a couple of "night" pictures from the Barrow sea-ice webcam in the past few days: the first, a beautiful scene, taken about an hour and a half before sunrise on Monday, and the second taken a little before solar midnight last night.

Monday, August 18, 2014

Barrow-Area Temperatures - Part 2

[Updated August 19 with polar plots, per reader request.]

A few readers of my original post about the Barrow urban heat island effect suggested that wind speed and direction are important in determining the relative temperatures at the Barrow ASOS and CRN stations.  This is undoubtedly true based on the Hinkel et al study, but I thought I could follow up with an analysis of the 12 years of overlapping data (Hinkel used only a single winter).  To do this, I calculated a daily mean wind vector from the surface observations in the twice-daily balloon soundings, then categorized the wind by both speed and direction and finally obtained the mean temperature difference between the two stations for each wind category.  (Note that a more accurate daily wind vector could be obtained from the hourly ASOS observations; the balloon soundings generally go up only at 3 am and 3 pm AKST.)

The results are shown below, for daily maximum and minimum temperatures in high summer (top 2 charts) and two months from deep winter (bottom 2 charts).  The blue, red, and green lines show the mean temperature differences (ASOS minus CRN) for weak (0-10 kt), moderate (10-20 kt), and strong (20+ kt) winds respectively.  I required at least 10 instances in a single category to compute a mean temperature difference, so there are large gaps in the results for higher (less common) wind speeds.  The purple columns indicate the frequency distribution of the wind direction categories; note that the categories are overlapping.

Based on this analysis, the most pronounced temperature differences between the Barrow ASOS and CRN stations are found for maximum temperatures in summer when winds are out of the east or east-southeast; in this scenario the trajectory of air reaching Barrow airport passes over a lot more land than the trajectory of air reaching the CRN station, and so the airport enjoys the benefit of more solar heating.  The difference is greatly lessened, however, when wind speeds increase, as expected.  When winds come from the west, there is little difference in summer temperatures between the two locations, and high temperatures are even a little lower at the airport with a stiff breeze from the west.

The daily minimum temperature differences for winter show a pattern that is nearly opposite to the high temperatures in summer, with temperature differences minimized for east or east-southeast flow, but with the airport staying a few degrees warmer when there is a light breeze from the west.  The maritime influence is greatly lessened in winter owing to sea ice cover, but it still plays a role in elevating the CRN temperatures when the trajectory is from the east or east-southeast.  Another way of looking at this is that we would normally expect Barrow temperatures to be noticeably higher in winter owing to the heat island effect, but when the flow is from the east or southeast, the air reaching the airport passes over more land and has time for radiative cooling, which brings the airport temperature down relative to the CRN location.

I'm sure there are several more interesting features of the data that could be discussed, but I'll leave it to readers to point these out and suggest hypotheses about their origin.

Update: reader Eric suggested using polar graphs to show the data; this is a great idea and I was able to get close to a true polar graph using the radar chart function in Excel.  See below, along with the station location map for ease of reference.

Saturday, August 16, 2014

Northwest Warmth

Kotzebue has been enjoying an extraordinary spell of warm late summer weather lately, and has broken some longstanding records in the process.  The first half of August was the warmest on record, by more than 1.5 °F, based on complete data since 1949 and patchy data back to 1929 (and some earlier years).  The previous record was exceeded for the daily mean temperature as well as for the August 1-15 average daily maxima and minima separately; the previous record was set in 1968 for all three of these metrics.  The chart below shows the temperature anomaly since June 1; until the recent heatwave developed, the summer was averaging a little cooler than normal.

In addition to the August 1-15 record, the week ending August 12 was the warmest week on record that was either wholly or partly within August, with a mean temperature of 66.2 °F compared to 65.7 °F for August 17-23, 2004.  The week ending August 14 also had the highest weekly average low temperature on record for a week ending in August (59.9 °F compared to 58.9 °F in 1977), but the weekly average high temperature was not quite as high as in the 2004 warm spell.  Also, the recent weekly average temperatures did not break the all-time records from June and July of previous years, although the week ending August 12 was the warmest week since early July of 1990.

Brian also points out via email and his Facebook page that the two-week period ending today in Kotzebue has had the warmest 14-day average high temperature since 1898 (at any time of year); so the warm spell is just about unprecedented in this respect.

And one more statistic: the number of days with high temperature of 70 °F or above is already a record for August (8 compared to 7 in 2004), and the number of nights with low temperature of 60 °F or above is easily a record for August (6 compared to only 3 in 1933).

It's interesting to look at the vertical profile of temperature anomalies from the twice-daily Kotzebue balloon soundings this summer, and to compare with the observations from the Barrow soundings about 300 miles to the north-northeast (see charts below).  At Kotzebue, the anomalies aloft have matched up very well with the surface anomalies, and have exceeded +8 °C at times recently.  The upper-air temperatures at Barrow have followed a similar path, with considerable warmth aloft in August, but interestingly the low-level temperatures have continued to track below-normal at Barrow this month.  This illustrates the "de-coupling" of the low-level air from the upper-level air that tends to persist at Barrow owing to the climatological inversion that remains all the way through late August (see discussion here).

Friday, August 15, 2014

Long-Term Precipitation Trends – Part I: Alaska

** Updated 8/17 with charts reflecting percentage changes instead of raw number changes **

Continuing with the theme of precipitation trends, I wanted to look at the rate of change of various precipitation thresholds over the last number of years to see if they are more or less frequent. To evaluate the rate of change, I looked at 4 different precipitation thresholds over the course of the Fairbanks climate record (good precipitation records extends back to 1915). Figure 1 shows the annual number of days for four different precipitation classes in Fairbanks (>= 0.01", >= 0.05", >= 0.25", and >= 0.50"). The only one of the trend lines that was significant at the 95% level (p<=0.05) was the number of days per year with at least 0.01" of precipitation. All the other thresholds were close to the even line but none showed a statistically significant trend (positive or negative). End of story, right?

Figure 1. Precipitation frequency (days per year) at Fairbanks for four different precipitation classes from 1915 to 2013. They are: 1) at least 0.01", 2) at least 0.05", 3) at least 0.25", and 4) at least 0.50".

A dangerous practice in the field of climatology is to make sweeping generalizations by looking at a single station's data. So, let's look to see how things have changed across a wide swath of Alaska. Unfortunately there are not many stations with a record as long as Fairbanks. We can increase the sample size to 21 stations when we set the begin date at 1951 instead of 1915. Figure 2 shows the stations we will use to look at the statewide trend in precipitation events.

Figure 2. Stations (21) used to assess long-term precipitation frequencies.

Looking at the statewide data, I was originally looking at "heavy" events so the categories are a little different. Actually, I was looking for national trends so the criteria were set somewhat higher than would ordinarily be appropriate for Alaska. In any event, the categories are: 1) at least 0.05", 2) at least 0.50", 3) at least 1.00", and 4) at least 2.00". That makes for a little bit of an apples to oranges comparison with Figure 1 – but let's progress nonetheless. Figures 3 through 6 below show the annual percentage difference of all stations from the 1951-2013 station average. To eliminate the effect of stations with large values overwhelming stations with low values, percentage differences were used instead of raw numbers. For example, a station that averages 50 days with 0.05" from 1951-2013 has a 10% increase if a year recorded 5 extra days with 0.05". A station that average 20 days with 0.05" from 1951-2013 has a 25% increase if a year recorded 5 extra days with 0.05". If you combine the raw numbers from those two stations, they would sow a 15% increase in the number of days with 0.05". However, comparing the percentages yields an increase of 17.5%. Therefore, comparing the annual percentage deviation is more meaningful since stations with larger averages would ordinarily drown out stations with low averages.

The number of days with >= 0.05" and >= 0.50" are noticeably increasing at a statistically significant rate (95% level) in Alaska. The >= 1.00" and >= 2.00" give mixed signals and neither trend is statistically significant. This is due in large part to small numbers and a relatively few stations responsible for most of the deviations. 

Therefore, I feel comfortable stating that the number of small precipitation events and the number of moderately-sized precipitation events have definitely increased over the last 60 years here in Alaska. This is not so surprising given the fact that warmer air can hold more moisture and warmer air at the surface with constant temperature aloft (not analyzed) would lead to lapse rate instability. As for the frequency of 1" and 2" precipitation events, they are too infrequent to make judgments on.

Figure 3. Statewide average of percentage change from 1951-2013 average annual number of days with at least 0.05" of precipitation.

Figure 4. Statewide average of percentage change from 1951-2013 average annual number of days with at least 0.50" of precipitation.

Figure 5. Statewide average of percentage change from 1951-2013 average annual number of days with at least 1.00" of precipitation.

Figure 6. Statewide average of percentage change from 1951-2013 average annual number of days with at least 2.00" of precipitation.

Wednesday, August 13, 2014

Hourly Temperature / Rainfall Numbers

One of the great data sets that NCDC archives is the hourly ASOS observations. These are particularly useful for developing climate descriptions and analyses on relatively short timescales. Much has been made of this summer's cool temperatures and excessive rains but little has been noted on these shorter time-scales.

I decided to compare last year's hourly observations with this year's hourly observations for the June 1 to August 10 time period. Instead of using the DS 3505 ISH data, I pulled the METAR observations directly. There are certain advantages to using the METARs instead of the DS 3505 data; e.g., the distinction between moderate and heavy rain.

From a temperature perspective, 2014 has been substantially cooler than 2013 in Fairbanks. Figure 1 shows the average hourly temperatures during the June 1 to August 10 time period for 2013 (orange) and 2014 (purple). The 25th percentile and 75th percentile lines are also shown. Interestingly, the warmest 25% of this summer's observations are about the same as the mean of last summer.

Figure 1. Average, 25th percentile, and 75th percentile temperatures for every hour of the day during the June 1 to August 10 time period for 2013 and 2014.

Of course this has been an exceptionally wet summer in Fairbanks. As we noted last week, the average day with precipitation has been quite wet. Looking at the hourly data, what we see is that not only have there been many more wet days this summer, but when it has rained, many more daily observations have indicated rain. Figure 2 shows some summary statistics for the number of rainfall observations. The last column on Figure 2 shows that days with rain have more than twice as many daily rainfall observations as last summer. In fact, 10 of the 71 days this summer have recorded rain for 12 or more of the 24 hourly observations. Last year, there were no such days.

Figure 2. Count of the number of rain observations during the June 1 to August 10 time period for 2013 and 2014.

Sunday, August 10, 2014

Wind Direction and Precipitation

Readers Eric and Gary have commented recently on the dependence of Fairbanks precipitation on wind direction.  It's widely understood (at least among weather enthusiasts) that westerly or southwesterly flow aloft tends to bring moisture to Fairbanks-land, but to what extent is this relationship borne out in the historical data?  This is a pretty straightforward analysis: I calculated the vector average wind speed at 700 mb (about 3 km MSL) from balloon soundings each day since 1948, and then I categorized the data into 30-degree direction windows, e.g. 0-30°, 10-40°, and so on.  For each direction window, and requiring a minimum wind speed of 10 knots, I then extracted the frequency of measurable precipitation and the average daily precipitation; the results are shown below.

As expected, the highest frequency of measurable precipitation is found for near-westerly flow at 700 mb; for summer the peak is west-southwesterly (230-260°), but for the other seasons, it's close to westerly.  All seasons also show a secondary peak in precipitation frequency; again the summer is distinct, with a frequency peak for easterly flow, but for the other seasons the secondary peak occurs for southeasterly flow.

The second chart, showing average daily precipitation amount for all days (not just non-zero precip days), also indicates a peak in precipitation amounts for westerly or west-southwesterly flow.  However, in summer the average amounts are nearly as high for northwesterly flow (310-340°) and for north-northeasterly flow.  The other seasons show much lower amounts in non-westerly flow regimes, although there is still a secondary peak for southeasterly flow in spring.  It's interesting to observe the extreme dryness of northerly and northeasterly flow in winter: there's no moisture to be had in that situation.

Finally, it's interesting to see the total contribution of each flow direction to the total precipitation at Fairbanks - see below.  The peak is pronounced in all seasons for flow just south of westerly (240-270°), because the preferred wind direction is out of the southwest; so even though other flow regimes can also be favorable for precipitation, they are less common and so contribute less to the total.  (The most common wind direction is 200-230° in autumn and winter, 190-220° in spring, and 230-260° in summer.)

Thursday, August 7, 2014

Fairbanks Precipitation Classes

Readers of the this blog do not need to be told that this has been an extraordinarily wet summer. This is already the second wettest June through August on record in Fairbanks. Are we nearing the wettest summer on record because is has rained often or because when it does rain, it rains a lot. To answer this question, let us look at all rainy days in the Fairbanks historical record. Figure 1 shows a series of lines representing the number of days with A) any precipitation, B) measurable precipitation, C) >= 0.10" precipitation, and D) >= 0.50" precipitation for all years during the June 1 to Aug 5 time period.

On average, there are 24 days during this 66-day window with measurable precipitation. In 2014, there have been 29 days between June 1 and August 5 with measurable precipitation (range: 10 to 53). If you include Trace amounts, the long-term average is 38 days and the 2014 value is 47 (range: 14 to 58). In both instances, 2014 is above the long-term average but quite far from the largest values. However, 2014 does lead in several important categories – days with >= 0.25", >=0.33", >=0.50", and >=1.00".

Figure 1. Number of days with A) any precipitation, B) measurable precipitation, C) over 0.10" precipitation, and D) over 0.50" precipitation for all years during the June 1 to Aug 5 time period.

This leads to the conclusion that when it rains, it pours. Figure 2 shows the June 1 to August 5 total precipitation (green line) and the per rainfall event amount (purple line). As you can see, 2014 has reported approximately 0.37" of rain per day with measurable rainfall. Only one other year (1962) even comes close to this year's per rainy day average. That year had 10 fewer days with measurable precipitation through August 5.

Figure 2. Total precipitation and average precipitation per rainy day for all years during the June 1 to Aug 5 time period.

Nine times since June 1st the Fairbanks daily precipitation has exceeded the 95th percentile for the date. In theory, during any 66-day period, the 95th percentile should be exceeded 3.3 times. Figure 3 shows the Fairbanks precipitation exceedance probabilities with 2014 precipitation events overlaid.

Figure 3. Daily precipitation exceedance probabilities with 2014 precipitation events overlaid. Note: The probability lines only take into account days with 0.01" or greater.