Ryan McGinnis
EF5
Interesting article in the WSJ this weekend:
http://online.wsj.com/article/SB123699110401126689.html#
Excerpt:
http://online.wsj.com/article/SB123699110401126689.html#
Excerpt:
WSJ: Study Links Tornado to Drought and Urban Heat
- Thread starter Ryan McGinnis
- Start date
Jeff Snyder
EF5
Hmm, I'm not so sure about this particular research project... You can view radar-estimated precip from the days leading up to that event at http://water.weather.gov/index.php ("archive:daily" for days leading up to 3/14/08), and it certainly doesn't appear that there was much precip - certainly not enough to make me think that there was a resulting Td gradient. With rather limited amts of rainfall in the preceeding days, how much evaporation really occurred, and to what extent did the evaporation affect low-level (not just surface, but lowest 75-100mb) mixing ratios? I don't entirely remember the event, or the week leading up to the tornado, but, at least from the archived data, I'm not confident that previous rainfall, as light as it apparently was, really resulted in any additional "boundaries". Perhaps if some areas received a lot of rainfall in the days leading up to 3/14/08 I'd feel a bit better... We've seen Td gradients (and, thus, probable Theta-e gradients) result from large gradients in precip accumulations during dry periods, but I don't know how deep that evaporated moisture really gets... It seems reasonable that it can mix to considerable height, but the mixing also "dilutes" this injection of moisture.
"Heat island" effects can result in higher temperature in the areas downwind of the source (e.g. Atlanta), primarily in the evening and overnight hours, but how big is the volume of air affected the urban temperature effects compared to the volume of air ingested by a supercell? It's obvious that surface temps can be affected by the heat island effect, but how much does it increase the mean 100-mb temp of a layer that's being ingested by a supercell? That's an honest question, fwiw. I can believe that some heat effects can influence deep, moist convection (e.g. a supercell), but was it an influence that "caused" this particular storm to produce a tornado?
All of this is to say that it's known that precip patterns are associated with severe weather occurrence (at least they're correlated, though it seems reasonable to me that there would be a cause-and-effect relationship - dry soils lead to greater sensible heating at the surface, which can intensify ridging aloft, which can further discourate precip, etc). However, did the light precip really "cause" or even affect this particular storm? Heat island, perhaps (influenced, however slightly, the evolution of the storm). Light precip in preceeding days, I think most likely not.
I await the paper, I guess.
"Heat island" effects can result in higher temperature in the areas downwind of the source (e.g. Atlanta), primarily in the evening and overnight hours, but how big is the volume of air affected the urban temperature effects compared to the volume of air ingested by a supercell? It's obvious that surface temps can be affected by the heat island effect, but how much does it increase the mean 100-mb temp of a layer that's being ingested by a supercell? That's an honest question, fwiw. I can believe that some heat effects can influence deep, moist convection (e.g. a supercell), but was it an influence that "caused" this particular storm to produce a tornado?
All of this is to say that it's known that precip patterns are associated with severe weather occurrence (at least they're correlated, though it seems reasonable to me that there would be a cause-and-effect relationship - dry soils lead to greater sensible heating at the surface, which can intensify ridging aloft, which can further discourate precip, etc). However, did the light precip really "cause" or even affect this particular storm? Heat island, perhaps (influenced, however slightly, the evolution of the storm). Light precip in preceeding days, I think most likely not.
I await the paper, I guess.
Last edited by a moderator:
David Wolfson
EF5
I continue being a curmugeon about the "widely accepted" principle that "wet ground breeds tornados." Apart from the Purdue study which is a debateable sample of one, whatever it says, I'd sure like to see a more rigorous justification based on atmospheric science rather than various sorts of statistical correlations. My question still is, what is/are the physical mechanisms relevant over the time and spatial scales that would affect supercell and tornado genesis?
The usual obvious starting-point -- that wet ground significantly increases surface moisture which in turn enhances severe storm development -- sounds good IMHO but may come up short in the science department. First, evapo-transpiration, by itself doesn't increase atmospheric instability as measured by near-surface Theta-e; rather it trades sensible for latent heat. And in fact it's a "widely accepted" principle of the Southwest summer desert monsoons that wet ground Day 1 inhibits storm formation on Day 2. The plausible physical explanation for this is that the lowered surface temperature adds CIN to the mix below the level of free convection.
Second, wet ground ideally only provides enough for a few days of significant evaporation. Sun and dry wind dessicate the surface soil in rather short order, which is why agriculture doesn't depend on it. Theories that rely on early Spring moisture enabling mid-Spring westward drylines and higher ambient moisture just don't add up for me.
Third, what does make some sense to me is a related but distinct mechanism. Dry, bare soil has a much lower albeido (higher heat reflectance) than well-vegetated ground. The effect of increased vegetation all-things-equal should sustain a significant increase in near-surface mesoscale atmospheric Theta-e relative to bare soil over a rather long time. Thanks to more normal winter season moisture the Arizona deserts are notably greener than during drought. The optical difference is significant and lasts for awhile, unlike surface moisture that's gone in a day or two.
I'd really appreciate someone who has a good scientific handle on the subject making a case for what the article asserts.
The usual obvious starting-point -- that wet ground significantly increases surface moisture which in turn enhances severe storm development -- sounds good IMHO but may come up short in the science department. First, evapo-transpiration, by itself doesn't increase atmospheric instability as measured by near-surface Theta-e; rather it trades sensible for latent heat. And in fact it's a "widely accepted" principle of the Southwest summer desert monsoons that wet ground Day 1 inhibits storm formation on Day 2. The plausible physical explanation for this is that the lowered surface temperature adds CIN to the mix below the level of free convection.
Second, wet ground ideally only provides enough for a few days of significant evaporation. Sun and dry wind dessicate the surface soil in rather short order, which is why agriculture doesn't depend on it. Theories that rely on early Spring moisture enabling mid-Spring westward drylines and higher ambient moisture just don't add up for me.
Third, what does make some sense to me is a related but distinct mechanism. Dry, bare soil has a much lower albeido (higher heat reflectance) than well-vegetated ground. The effect of increased vegetation all-things-equal should sustain a significant increase in near-surface mesoscale atmospheric Theta-e relative to bare soil over a rather long time. Thanks to more normal winter season moisture the Arizona deserts are notably greener than during drought. The optical difference is significant and lasts for awhile, unlike surface moisture that's gone in a day or two.
I'd really appreciate someone who has a good scientific handle on the subject making a case for what the article asserts.
Last edited by a moderator:
Greg Campbell
EF5
Did anyone see a link to the actual publication?
The WSJ article has the ring of inaccuracy and gives the impression that the Purdue findings are losing something in the translation.
The WSJ article has the ring of inaccuracy and gives the impression that the Purdue findings are losing something in the translation.
terra seright
I'm trying to remember how dry it was in the TX panhandle prior to the 2007 season, LOL. I've been doing a rain dance here, hoping for enough moisture! I was pondering this subject in the past couple of weeks, and I was thinking it was pretty dry, then storm season kicked in and we got mass rain/storms/tornadoes. Could be wrong, it's happened once before. I don't have any idea how to get that particular info-maybe one of you edumakated meteorlogists can find out? Sure seems like we had a dry year in summer/fall/winter 2006, with cracks in the ground that you could put a 20' 2x4 in and lose it....then storm season hit in March and in April of 2007 we had a freeze, then all hell broke loose and we had 50-something during one season...??
Jay McCoy
EF5
Its hard to correlate precip measured at the NWS office with the tornado count for the entrie panhandle. We had zero tornados at the NWS office..lol . Its better to look at overall panhandle rainfall averages but here is their review data from 2006 and 2007 anyway.
We were in a major drought in 2006 and had those devastating wildfires in march that burned 1 million acres and killed 12 people (biggest wildfire in texas history) but the dry spell ended in July/Aug when we had over 11 inches in those 2 months and continued into a wet fall and ended up 2" above normal for the year.
2007 started off near normal until march when we had over 4" and 19 tornados. April was wierd as we only recorded .65 of precip at the NWS but had 17 tornados for the month. But by the end of June we had 12.5" of rain (above normal) and 63 tornados.
data from ama NWS
http://www.srh.noaa.gov/ama/climate/2006wxreview.htm
http://www.srh.noaa.gov/ama/climate/2007wxreview.htm
Ofcourse now we are in another major drought. if we dont get some serious rain soon the dryline will be along I-35 for most of the season. I cant comment on how much ground moisture helps increase tornado activity or strength but I can talk about how much it limits the dryline mixing east on a given day. The drier the ground here in the panhandle the more the dryline mixes east...period.. Have seen it may times over the years. Evapotransportation is key to where the dryline will set up on a given day.
Last edited by a moderator:
Brandon Lawson
EF2
I agree that dryer surface conditions will increase the rate at which a dryline mixes eastward, but isn't the initial location of the dryline largely dependent on where the surface low develops, which is largely dependent on the strength and placement of the trough?
My question would be how do dry conditions affect the overall upper-level pattern? It sounds like most people are of the mind that drought conditions lead to ridging. If the conditions were dry enough, would it be possible to end up with a thermal low instead, or would that require some extreme desert-like heating? Not that it would change any of the moisture problems.
My question would be how do dry conditions affect the overall upper-level pattern? It sounds like most people are of the mind that drought conditions lead to ridging. If the conditions were dry enough, would it be possible to end up with a thermal low instead, or would that require some extreme desert-like heating? Not that it would change any of the moisture problems.
Jay McCoy
EF5
Partly but its hard to get deep moisture west when the ground is so dry. The dryline tends to develope further east close to the edge of the caprock when the panhandle is dry. Any moisture is literally sucked out of the air as it moves west up onto the caprock but the moisture does pool at the edge of the caprock. You can have a trough further west but the moisture is very shallow and it isnt an actual dryline until it moves east and catches up with the moisture. That is a big problem we have in dry years is we have the low trough west but no real moisture and storms wont really get going until it slams into deeper moisture east. Alot of high based LP's and virga but once it hits the edge of the caprock and the deeper moisture they explode but its usually getting closer to dark by then so chasing is crappy.
