Saturday, November 1, 2014

Historical Cloud Cover

In recent weeks I put up a few posts looking at changes in mean seasonal temperature since 1950 in Fairbanks, with a particular focus on winter warming; the most recent post was here.  One of the obvious next steps in the investigation is to look at changes in winter cloud cover over time, because there is a strong relationship between cloud cover and temperature (see e.g. here).

I plan to look at this more fully at another time (I'm currently traveling and unable to write a detailed analysis), but for now I'll put up a couple of charts and invite comment.  First, the fraction of the hourly sky condition observations that reported "clear" in deep winter (Dec-Feb) since 1950, see below.


One issue to be aware of here is that the automated (ASOS) observations were introduced in 1997, and it's inevitable that this change produced a bias relative to earlier years, because the ASOS ceilometer doesn't detect clouds above a certain height (12000 feet in the original ASOS ceilometer).  For example, here's a comparison of ASOS and conventional observations for several stations in the lower 48 over a limited period of time (taken from a report here).

These results indicate, as we would expect, that the ASOS instrument over-reports clear skies compared to earlier manual observations, so if anything we would expect to see a higher frequency of clear skies since 1997 in the chart above.  It seems that the change in observing procedure does not explain the apparent decrease in frequency of clear skies in recent winters in Fairbanks.

Another potentially confounding factor is that fog in winter in Fairbanks obscures the sky and precludes cloud cover observations, and we know that fog used to be more common in Fairbanks (see e.g. here).  However, we would expect that fog forms preferentially under clear skies in winter, so presumably the missing observations from foggy times in earlier years would be more of the "clear sky" variety.  Again, the trend towards less fog (more complete cloud cover data) doesn't seem to explain the decrease in frequency of clear skies.

Another way to look at the data is to calculate the mean cloud cover fraction from the reported coverage category (i.e. clear, scattered, broken, or overcast); see below.  I hope to post more discussion and analysis at a later date.


*******************
Charts added by Brian on 11/2 & 11/3




21 comments:

  1. So what I gather, hopefully correctly, is that the number of clear skies have been steadily decreasing over the last 30 years. However, the mean cloud cover has been constant. So the number of fully cloudy days has also decreased. Which means more partly cloudy readings. Which means less rain and snow storms in general - which we have seen. A general increase in clouds would also increase winter temps and decrease ice fog.

    ReplyDelete
    Replies
    1. Eric, I'll look more fully at the complete distribution of cloud cover changes in the coming days. It does seem there have been some interesting changes.

      Delete
  2. This comment has been removed by a blog administrator.

    ReplyDelete
  3. I'll finish the ASOS vs Observer analysis tomorrow, but for now this says quite a bit about the function of machine vs person:

    "While improving the efficiency in acquiring weather data, the ASOS system lacks the observational ability of the human observer to spatially integrate some of the weather elements over a large area. Two such elements are ceiling height and opaque cloud cover, which are important in estimating atmospheric stability and mixing heights required for applications of several regulatory and non regulatory dispersion models."

    And: SNIP..."This view also allows the observer to modify the ceiling height if there are high clouds, and to make a determination on the transparency of the clouds."

    All of this machination appears to have compromised the database above the 12,000 deck.

    Gary

    ReplyDelete
  4. Another thought: could warmer surface water in the Gulf of Alaska in winter lead to deeper storms which take longer to dissipate? Stronger storms would more effectively mix the atmosphere over the gulf/ mainland/ Bearing sea circuit. A stronger Aleution low would lead to more southerly flow over the mainland. A good question has the mean position of the winter Aleution low drifted over the past 50 years. Thoughts?

    ReplyDelete
  5. Richard, I took the liberty of adding a chart to your excellent post.

    Earlier this year I was looking into the same phenomenon down here in Anchorage and noticed that the number of "Clear" sky observations were few and far between. In fact, there were days when not a single cloud was present all day but many of the hourly observations reported "Few" for the sky condition.

    Counterintuitively, the number of "Overcast" observations has also decreased. As you noted, the most likely culprit is the transition from manual to ASOS observations. Since the annual percent cloud coverage is pretty constant, the difference must be due to the observation technique.

    ReplyDelete
    Replies
    1. The Anchorage graph seems pretty clear that a change in procedure caused such a huge change in cloud cover measurements. But I would argue that there is still a general trend towards cloudiness overall for both Anchorage and especially Fairbanks. Which accords with what we know about the general climatology.

      Delete
    2. Eric, I added a second chart for Fairbanks. With the advent of the ASOS, there seems to be A) fewer Clear observations, B) fewer Overcast observations, C) more Few sky observations, and, D) far more Broken sky observations.

      Delete
  6. I assume the ASOS looks straight up for any clouds. Unless it scans the celestial dome like a human, it'd present a limited view of what's overhead.

    The machine also depends on the lateral movement of clouds to establish an averaged value of overcast conditions over a pre-determined time interval. Any change in rate of movement would influence the product.

    Gary

    ReplyDelete
  7. Are the ASOS obs trends of sky condx consistent for Interior Alaska? Are the trends a function of machine obs or broad scale weather?

    Gary

    ReplyDelete
    Replies
    1. This reads confusing even to me. I meant to ask: Have other/all ASOS stations in Interior Alaska reported similar trends in cloud cover as Fairbanks? If so, can we assume it's a function of man vs machine?

      PS: Clear today to my eyes...FAI ASOS reports thin clouds as FEW 100. Must be hitting an invisible layer.

      Gary

      Delete
  8. This is lunacy. The ASOS is reporting "A Few Clouds FEW070 FEW130" I just walked outside and there's not a cloud in the sky for 360 deg. Severe clear iId call it.

    Gary

    ReplyDelete
  9. Thanks everyone for the comments, and Brian for the extra charts. I was traveling and busy until today, so unable to contribute to the discussion.

    There's no doubt the ASOS introduced significant biases relative to earlier manual observations. However, I'm not sure all the changes in recent years are spurious. In Anchorage, the ASOS was introduced on June 1, 1998 according to the NCDC metadata, but the cloud obs showed big changes in 1996 and 1997. In Fairbanks the change is listed as occurring on December 1, 1997, but the big spike in "FEW" observations was for all of 1997.

    It's unfortunate that the super-El Nino of 1997-1998 occurred in the same time frame as the ASOS transition, as the Pacific anomaly would be expected to affect cloud cover over Alaska.

    Gary - I'll see what I can find for other stations. This issue is an important one, given the tremendous importance of cloud cover for climate in general and winter temperatures in particular.

    ReplyDelete
    Replies
    1. Richard it's been another clear day in Fairbanks. I was at the airport from 23-2400Z today and there were no visible clouds = clear as it ever gets. The ASOS is still reporting FEW @ 10-12K.

      ASOS: http://www.nws.noaa.gov/asos/

      From The Trainer's ToolKit above: "At rare times, ASOS may report a dense moisture layer as opaque clouds before the layer becomes totally visible to the human eye."

      Today's Skew-T/log-P plot for 700 mb does show a wind shift and dew point change. Maybe that's what it's recording?

      Regardless of the source, I suggest that with the current ASOS condition "Few" be integrated with "Clear" until the machine's calibration is confirmed. I'm skeptical of it's reports of Few versus visual obs for those altitudes. Maybe the NWS has a better idea what's happening.

      Gary

      Delete
    2. Today was clear because of high pressure creating an inversion. Instead of a water vapor layer perhaps the ASOS laser was picking up a distinct layer of smog. During winter there is always a cloud of smoke from smokestacks over near the airport entrance. Some of that may have drifted over above the sensor.

      Maybe if we compare ASOS cloudiness, inversion strength, and the borough's official air quality measurements we might see a detectable correlation.

      Delete
    3. Good point Eric. We also get that Asia-derived Arctic Haze that drifts in from that direction. At the altitudes the ASOS is reporting clouds I doubt the inversion would allow much local particulate to rise that high. I've flown through layers of that haze. It's typically several thousand feet AGL and brown in color.

      The ASOS did call "CLR" on a few obs last eve so it's capable of doing that. Something's up there that it's detecting. Wonder what it is?

      Gary

      Delete
  10. The last couple of nights (11/5-6, 11/6-7/14) the Fairbanks ASOS has reported CLR skies from approximately 7pm to 7am. That's after a daylight period of reporting FEW when my ground obs fail to detect visible clouds.

    Could the low angle of the Sun this time of year (8-9 deg) be interacting with ice crystals (or ?) at altitude in such a way as to fool the ASOS into detecting a cloud layer?

    Why would CLR be associated with dark, and Few be associated with daylight periods on otherwise visibly clear days?

    Gary

    ReplyDelete
  11. Gary, that is a very interesting observation and a quick look at the data since 2000 indicates a fairly pronounced diurnal cycle in the frequency of CLR during winter. For December-February since 2000, the frequency of CLR peaks near 17% at around 11-14 UTC, i.e. 2-5 am AKST, and bottoms out near 5% at around 19-23 UTC, i.e. 10am-2pm AKST.

    However the frequency of FEW is also minimized around noon local time (but with much smaller differences between hours), so it is not a simple matter of switching between the two.

    Interestingly there is also a similar diurnal cycle (but with much higher percentages of CLR) in earlier decades, so this is not solely an ASOS artifact.

    Obviously this topic is worthy of additional investigation.

    ReplyDelete
    Replies
    1. I can see where the darkness of winter might make human observations of clouds at altitude (especially thin wispy clouds) a challenge. Maybe they look for discontinuities in the nightly celestial dome, use vertical lights, send up balloons, incantate.

      But the ceilometer's laser -driven backscatter should work regardless of time of day. That's unless something is confusing the instrument's input, as it appears to me.

      As to whether or not reports of FEW or CLR have diurnal singularities that vary with season, more will have to be revealed.

      I suppose a trip to the NWS for me for an ASOS briefing is in order.

      Gary

      Delete
  12. Hi there,

    Would you be interested in my study about the (perceived?) change in historical cloud cover in Nashville, TN?

    ReplyDelete