From dshelms at comcast.net Wed Nov 2 21:29:51 2005 From: dshelms at comcast.net (Dave Helms) Date: Wed Nov 2 21:32:41 2005 Subject: [wxqc] On the Calibration of the David Tipping Bucket Rain Gauge... In-Reply-To: <102120052053.28104.435955370007308E00006DC822028887449C03040A089C0B@comcast.net> References: <102120052053.28104.435955370007308E00006DC822028887449C03040A089C0B@comcast.net> Message-ID: <4369761F.6010301@comcast.net> Hi All: This post comes by way of Andy Overton/UK Weather Observer Email: dormergreen at tiscali.co.uk On the subject of calibration, recent correspondence amongst members of the UK Climatological Observers Link reveals a consistent problem with the calibration of the Davis raingauge "out of the box". As a fair proportion of CWOP members are probably using this system it might be worth investigating as you could be getting erroneous data from these people. Everyone I have contacted here and who has done a calibration check has found their gauge under-reading by a consistent ~10-15%. Some who operated earlier Davis AWS's like Weather Monitor II had the same problem with those. Calibration has been of two kinds- by pouring a known amount of water in (the methodology is described on my website http://myweb.tiscali.co.uk/doncastertownmoorwx by clicking on the /Station Maintenance and Calibration/ page) and by comparison with a standard Snowdon pattern manual gauge. Both methods produce the under-read. I contacted Davis about this (we did wonder whether US water was heavier than UK water, which would explain it!) and they stated that they don't calibrate the gauges individually. They do one every so often and note the screw settings underneath the buckets, then set all others to this height. Something isn't working at all well with this system. Calibration is a bit time consuming but not intrinsically difficult with this gauge. If our results are repeated across the US (and World) this constitutes quite a problem for CWOP. Observers really need to calibrate those gauges- once done it shouldn't need doing again. We've also found that the calibration of the barometer from the factory isn't so good but that's easily corrected and CWOP cover this already in your notes to observers. Hope this is useful. Additional Note from DaveH: Good reason to have a 4" manual gauge as a comparison to the automated gauge. GaryO lists several rain gauge vendors on his Equipment page: http://www.weatherforum.net/sources.html From paulgrace at lookoutranch.com Sun Nov 20 23:13:05 2005 From: paulgrace at lookoutranch.com (Paul Grace) Date: Sun Nov 20 23:13:11 2005 Subject: [wxqc] Temperature error Message-ID: <000901c5ee51$df738a20$0a00a8c0@Dell1500> The CWOP info for my station (CW0622) is showing an ?error? of -4.1 ?F, the worst average error of -16.4. In looking at the neighboring stations, they are all in very different areas? I am on a 640m ridge 10 miles from Pt. Arena, California, while my ?neighbors? are 20 ? 60 miles east, in warmer valleys a few meters above sea level. I would expect lower temperatures at my station. I have a Davis fan-aspirated temp. sensor, that seems correct to within a fraction of a degree when compared to a thermocouple. It is situated in the middle of a small field, about 15ft off the ground, a couple hundred feet from the nearest trees. How are the ?correct? temperature readings interpolated/extrapolated/measured? Is my station being compared to these ?neighbors?? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://pond1.gladstonefamily.net:8080/pipermail/wxqc/attachments/20051120/8be34503/attachment.html From dshelms at comcast.net Mon Nov 21 00:45:53 2005 From: dshelms at comcast.net (Dave Helms) Date: Mon Nov 21 00:45:51 2005 Subject: [wxqc] Temperature error In-Reply-To: <000901c5ee51$df738a20$0a00a8c0@Dell1500> References: <000901c5ee51$df738a20$0a00a8c0@Dell1500> Message-ID: <43815F11.5080906@comcast.net> Hi Paul, MADIS QCMS uses a potential temperature comparison which corrects for elevation differences (dry adiabatic lapse rate) to assess station performance. Mesowest does a similar temperature comparison methodology, and also appears to show your temperatures running a little warm, after correcting for elevation: http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 We have seen that valley stations run a bit "cool", and mountaintop stations, such as yours, run a bit "warm" in these statistical analyzes. We believe these warm/cool bias anomalies appear in the statistics because the algorithm does not account the amount of atmospheric mixing occurring at individual sites. Valley sites tend to "de-couple" at night with little atmospheric mixing and added cooling from radiational cooling and drainage winds. Mountaintop stations are well mixed throughout the night and don't typically have evening radiational cooling to the degree seen in valley stations. Given that you have an active aspirated temperature sensor sited in an open area, I think that your temperature observations are representative for your location. A little nit... I would lower the elevation from 15 ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a big difference in performance, but that is the standard). The rain gauge should be similarly sited AGL to minimize potential under-catch caused by turbulent wind flow across the top of the rain gauge from stronger winds at the higher siting elevation. Dave CW0351 Paul Grace wrote: > The CWOP info for my station (CW0622) is showing an ?error? of -4.1 > ?F, the worst average error of -16.4. > > In looking at the neighboring stations, they are all in very different > areas? I am on a 640m ridge 10 miles from Pt. Arena, California, while > my ?neighbors? are 20 ? 60 miles east, in warmer valleys a few meters > above sea level. I would expect lower temperatures at my station. I > have a Davis fan-aspirated temp. sensor, that seems correct to within > a fraction of a degree when compared to a thermocouple. It is situated > in the middle of a small field, about 15ft off the ground, a couple > hundred feet from the nearest trees. > > How are the ?correct? temperature readings > interpolated/extrapolated/measured? Is my station being compared to > these ?neighbors?? > >------------------------------------------------------------------------ > >_______________________________________________ >wxqc mailing list >wxqc@lists.gladstonefamily.net >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > >The contents of this message are the responsibility of the author. > From david at coursey.com Mon Nov 21 01:32:23 2005 From: david at coursey.com (David Coursey) Date: Mon Nov 21 08:45:38 2005 Subject: [wxqc] Temperature error Message-ID: <2AA5D5B18E1F4849A2B998B43BAA2BA70FC6FA@fjuckby.coursey.local> I have a similar problem here at AR987 N5FDL in Tracy, Calif. My temps seem way off sometimes, though when I check them with another thermometer they always check out. I am in the Central Valley (by a few mines, but across the Altamont Range) some of the stations that I think I am being compared to are back toward the Bay Area and on the other side of the hills. Could someone take a look at my temps and tell me if they look OK? I am concerned that I am sending bad data, even though I am pretty sure I am not. My station is a Davis Vantage Pro II mounted 5-6 feet off a concrete tile roof at about 32 feet AGL. Thanks, David, N5FDL -----Original Message----- From: wxqc-bounces@lists.gladstonefamily.net [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of Dave Helms Sent: Sunday, November 20, 2005 9:46 PM To: Discussion of weather data quality issues Subject: Re: [wxqc] Temperature error Hi Paul, MADIS QCMS uses a potential temperature comparison which corrects for elevation differences (dry adiabatic lapse rate) to assess station performance. Mesowest does a similar temperature comparison methodology, and also appears to show your temperatures running a little warm, after correcting for elevation: http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 We have seen that valley stations run a bit "cool", and mountaintop stations, such as yours, run a bit "warm" in these statistical analyzes. We believe these warm/cool bias anomalies appear in the statistics because the algorithm does not account the amount of atmospheric mixing occurring at individual sites. Valley sites tend to "de-couple" at night with little atmospheric mixing and added cooling from radiational cooling and drainage winds. Mountaintop stations are well mixed throughout the night and don't typically have evening radiational cooling to the degree seen in valley stations. Given that you have an active aspirated temperature sensor sited in an open area, I think that your temperature observations are representative for your location. A little nit... I would lower the elevation from 15 ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a big difference in performance, but that is the standard). The rain gauge should be similarly sited AGL to minimize potential under-catch caused by turbulent wind flow across the top of the rain gauge from stronger winds at the higher siting elevation. Dave CW0351 Paul Grace wrote: > The CWOP info for my station (CW0622) is showing an "error" of -4.1 > ?F, the worst average error of -16.4. > > In looking at the neighboring stations, they are all in very different > areas- I am on a 640m ridge 10 miles from Pt. Arena, California, while > my "neighbors" are 20 - 60 miles east, in warmer valleys a few meters > above sea level. I would expect lower temperatures at my station. I > have a Davis fan-aspirated temp. sensor, that seems correct to within > a fraction of a degree when compared to a thermocouple. It is situated > in the middle of a small field, about 15ft off the ground, a couple > hundred feet from the nearest trees. > > How are the "correct" temperature readings > interpolated/extrapolated/measured? Is my station being compared to > these "neighbors"? > >----------------------------------------------------------------------- >- > >_______________________________________________ >wxqc mailing list >wxqc@lists.gladstonefamily.net >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > >The contents of this message are the responsibility of the author. > _______________________________________________ wxqc mailing list wxqc@lists.gladstonefamily.net http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc The contents of this message are the responsibility of the author. e responsibility of the author. From paulgrace at lookoutranch.com Mon Nov 21 12:33:12 2005 From: paulgrace at lookoutranch.com (Paul Grace) Date: Mon Nov 21 12:33:22 2005 Subject: [wxqc] Temperature error In-Reply-To: <43815F11.5080906@comcast.net> Message-ID: <002f01c5eec1$a8eefaa0$0a00a8c0@Dell1500> Thanks! 5ft will help in cleaning too- -----Original Message----- From: wxqc-bounces@lists.gladstonefamily.net [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of Dave Helms Sent: Sunday, November 20, 2005 21:46 To: Discussion of weather data quality issues Subject: Re: [wxqc] Temperature error Hi Paul, MADIS QCMS uses a potential temperature comparison which corrects for elevation differences (dry adiabatic lapse rate) to assess station performance. Mesowest does a similar temperature comparison methodology, and also appears to show your temperatures running a little warm, after correcting for elevation: http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 We have seen that valley stations run a bit "cool", and mountaintop stations, such as yours, run a bit "warm" in these statistical analyzes. We believe these warm/cool bias anomalies appear in the statistics because the algorithm does not account the amount of atmospheric mixing occurring at individual sites. Valley sites tend to "de-couple" at night with little atmospheric mixing and added cooling from radiational cooling and drainage winds. Mountaintop stations are well mixed throughout the night and don't typically have evening radiational cooling to the degree seen in valley stations. Given that you have an active aspirated temperature sensor sited in an open area, I think that your temperature observations are representative for your location. A little nit... I would lower the elevation from 15 ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a big difference in performance, but that is the standard). The rain gauge should be similarly sited AGL to minimize potential under-catch caused by turbulent wind flow across the top of the rain gauge from stronger winds at the higher siting elevation. Dave CW0351 Paul Grace wrote: > The CWOP info for my station (CW0622) is showing an ?error? of -4.1 > ?F, the worst average error of -16.4. > > In looking at the neighboring stations, they are all in very different > areas? I am on a 640m ridge 10 miles from Pt. Arena, California, while > my ?neighbors? are 20 ? 60 miles east, in warmer valleys a few meters > above sea level. I would expect lower temperatures at my station. I > have a Davis fan-aspirated temp. sensor, that seems correct to within > a fraction of a degree when compared to a thermocouple. It is situated > in the middle of a small field, about 15ft off the ground, a couple > hundred feet from the nearest trees. > > How are the ?correct? temperature readings > interpolated/extrapolated/measured? Is my station being compared to > these ?neighbors?? > >------------------------------------------------------------------------ > >_______________________________________________ >wxqc mailing list >wxqc@lists.gladstonefamily.net >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > >The contents of this message are the responsibility of the author. > _______________________________________________ wxqc mailing list wxqc@lists.gladstonefamily.net http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc The contents of this message are the responsibility of the author. From dshelms at comcast.net Mon Nov 21 14:20:13 2005 From: dshelms at comcast.net (dshelms@comcast.net) Date: Mon Nov 21 14:20:10 2005 Subject: [wxqc] Temperature error Message-ID: <112120051920.28437.43821DED000673F000006F1522058864429C03040A089C0B@comcast.net> FYI... Just to do a little Weather 101 (apologies in advance for the techno stuff) I probably simplified the situation along the northern California coast in my previous discussion. I was "slicing" the temperatures in your area this morning using MADIS mesonet page and noticed stations above 1,500 ft (500 meters) AMSL were in the 60's F (15-ish degrees C) while the stations below 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). A west coast weather guy would recognize that as a classic subsidence inversion (temperatures warm with height), but I am an east coast weather weanie (and a little slow, I'm afraid). Checking out today's Oakland, CA, sounding from the NWS weather balloon (aka radiosonde): http://weather.unisys.com/upper_air/skew/skew_KOAK.html FYI... the left-hand white line is the dew point trace, where the temperature and dew point converge, its wet; where they diverge, its dry. I'm assuming you are often looking down at lots of fog and low clouds from your ranch as a result of the subsidense inversion you live in/near. The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" T), you see the right hand white line (atmospheric temperature) starts around 1000 mb (177 m or meters) at about 8 C degrees and increases with height until it tops out at almost 20 C at about 800 meters (about 2,400 ft). So... more than decoupling, your station (and others between 2-3,000 ft near you) trace the warm air near the maximum temperature of the subsidence inversion! Again, MADIS QCMS assumes the atmosphere cools with height which is not true for your location in many cases. Thus, the inadvertent "flag" of your temperature. Mystery solved! Dave > Hi Paul, > > MADIS QCMS uses a potential temperature comparison which corrects for > elevation differences (dry adiabatic lapse rate) to assess station > performance. Mesowest does a similar temperature comparison methodology, > and also appears to show your temperatures running a little warm, after > correcting for elevation: > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 > > We have seen that valley stations run a bit "cool", and mountaintop > stations, such as yours, run a bit "warm" in these statistical analyzes. > We believe these warm/cool bias anomalies appear in the statistics > because the algorithm does not account the amount of atmospheric mixing > occurring at individual sites. Valley sites tend to "de-couple" at night > with little atmospheric mixing and added cooling from radiational > cooling and drainage winds. Mountaintop stations are well mixed > throughout the night and don't typically have evening radiational > cooling to the degree seen in valley stations. > > Given that you have an active aspirated temperature sensor sited in an > open area, I think that your temperature observations are representative > for your location. A little nit... I would lower the elevation from 15 > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a > big difference in performance, but that is the standard). The rain gauge > should be similarly sited AGL to minimize potential under-catch caused > by turbulent wind flow across the top of the rain gauge from stronger > winds at the higher siting elevation. > > Dave > CW0351 > > > > Paul Grace wrote: > > > The CWOP info for my station (CW0622) is showing an “error” of -4.1 > > °F, the worst average error of -16.4. > > > > In looking at the neighboring stations, they are all in very different > > areas— I am on a 640m ridge 10 miles from Pt. Arena, California, while > > my “neighbors” are 20 – 60 miles east, in warmer valleys a few meters > > above sea level. I would expect lower temperatures at my station. I > > have a Davis fan-aspirated temp. sensor, that seems correct to within > > a fraction of a degree when compared to a thermocouple. It is situated > > in the middle of a small field, about 15ft off the ground, a couple > > hundred feet from the nearest trees. > > > > How are the “correct” temperature readings > > interpolated/extrapolated/measured? Is my station being compared to > > these “neighbors”? > > > >------------------------------------------------------------------------ > > > >_______________________________________________ > >wxqc mailing list > >wxqc@lists.gladstonefamily.net > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > > >The contents of this message are the responsibility of the author. > > > > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. From paulgrace at lookoutranch.com Mon Nov 21 16:37:49 2005 From: paulgrace at lookoutranch.com (Paul Grace) Date: Mon Nov 21 16:38:20 2005 Subject: [wxqc] Temperature error In-Reply-To: <112120051920.28437.43821DED000673F000006F1522058864429C03040A089C0B@comcast.net> Message-ID: <001801c5eee3$db160f10$0a00a8c0@Dell1500> Yes, we do look out on a river valley often full of fog, we are almost always above it. So you're a "slow" weather weenie? Scary. For a big picture question; is the fact that QCMS tags my temps/dewpoint as bad make any difference with the use of my data? Is my data used for any actual purpose? There are very few stations near mine, so I thought it might be a different data point, but if it's tossed out, then should I bother uploading it? -----Original Message----- From: wxqc-bounces@lists.gladstonefamily.net [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of dshelms@comcast.net Sent: Monday, November 21, 2005 11:20 To: Discussion of weather data quality issues; Discussion of weather data quality issues Subject: Re: [wxqc] Temperature error FYI... Just to do a little Weather 101 (apologies in advance for the techno stuff) I probably simplified the situation along the northern California coast in my previous discussion. I was "slicing" the temperatures in your area this morning using MADIS mesonet page and noticed stations above 1,500 ft (500 meters) AMSL were in the 60's F (15-ish degrees C) while the stations below 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). A west coast weather guy would recognize that as a classic subsidence inversion (temperatures warm with height), but I am an east coast weather weanie (and a little slow, I'm afraid). Checking out today's Oakland, CA, sounding from the NWS weather balloon (aka radiosonde): http://weather.unisys.com/upper_air/skew/skew_KOAK.html FYI... the left-hand white line is the dew point trace, where the temperature and dew point converge, its wet; where they diverge, its dry. I'm assuming you are often looking down at lots of fog and low clouds from your ranch as a result of the subsidense inversion you live in/near. The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" T), you see the right hand white line (atmospheric temperature) starts around 1000 mb (177 m or meters) at about 8 C degrees and increases with height until it tops out at almost 20 C at about 800 meters (about 2,400 ft). So... more than decoupling, your station (and others between 2-3,000 ft near you) trace the warm air near the maximum temperature of the subsidence inversion! Again, MADIS QCMS assumes the atmosphere cools with height which is not true for your location in many cases. Thus, the inadvertent "flag" of your temperature. Mystery solved! Dave > Hi Paul, > > MADIS QCMS uses a potential temperature comparison which corrects for > elevation differences (dry adiabatic lapse rate) to assess station > performance. Mesowest does a similar temperature comparison methodology, > and also appears to show your temperatures running a little warm, after > correcting for elevation: > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 > > We have seen that valley stations run a bit "cool", and mountaintop > stations, such as yours, run a bit "warm" in these statistical analyzes. > We believe these warm/cool bias anomalies appear in the statistics > because the algorithm does not account the amount of atmospheric mixing > occurring at individual sites. Valley sites tend to "de-couple" at night > with little atmospheric mixing and added cooling from radiational > cooling and drainage winds. Mountaintop stations are well mixed > throughout the night and don't typically have evening radiational > cooling to the degree seen in valley stations. > > Given that you have an active aspirated temperature sensor sited in an > open area, I think that your temperature observations are representative > for your location. A little nit... I would lower the elevation from 15 > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a > big difference in performance, but that is the standard). The rain gauge > should be similarly sited AGL to minimize potential under-catch caused > by turbulent wind flow across the top of the rain gauge from stronger > winds at the higher siting elevation. > > Dave > CW0351 > > > > Paul Grace wrote: > > > The CWOP info for my station (CW0622) is showing an ?error? of -4.1 > > ?F, the worst average error of -16.4. > > > > In looking at the neighboring stations, they are all in very different > > areas? I am on a 640m ridge 10 miles from Pt. Arena, California, while > > my ?neighbors? are 20 ? 60 miles east, in warmer valleys a few meters > > above sea level. I would expect lower temperatures at my station. I > > have a Davis fan-aspirated temp. sensor, that seems correct to within > > a fraction of a degree when compared to a thermocouple. It is situated > > in the middle of a small field, about 15ft off the ground, a couple > > hundred feet from the nearest trees. > > > > How are the ?correct? temperature readings > > interpolated/extrapolated/measured? Is my station being compared to > > these ?neighbors?? > > > >------------------------------------------------------------------------ > > > >_______________________________________________ > >wxqc mailing list > >wxqc@lists.gladstonefamily.net > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > > >The contents of this message are the responsibility of the author. > > > > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. _______________________________________________ wxqc mailing list wxqc@lists.gladstonefamily.net http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc The contents of this message are the responsibility of the author. From dshelms at comcast.net Mon Nov 21 17:15:01 2005 From: dshelms at comcast.net (dshelms@comcast.net) Date: Mon Nov 21 17:15:06 2005 Subject: [wxqc] Temperature error Message-ID: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> Hi Paul, Your data are not really flagged as bad, just something that needs greater attention to see what the real deal is. Because there are very few observations from your area is a great reason for you to continue to contributing your observations. Ultimately, the local weather guys down at Monterrey and up in Eureka know from looking at the station data what is good and not so good. Again, MADIS QCMS is a first guess and not a final statement of quality. Cheers, Dave > Yes, we do look out on a river valley often full of fog, we are almost > always above it. > So you're a "slow" weather weenie? Scary. > > For a big picture question; is the fact that QCMS tags my temps/dewpoint as > bad make any difference with the use of my data? Is my data used for any > actual purpose? There are very few stations near mine, so I thought it > might be a different data point, but if it's tossed out, then should I > bother uploading it? > > -----Original Message----- > From: wxqc-bounces@lists.gladstonefamily.net > [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of > dshelms@comcast.net > Sent: Monday, November 21, 2005 11:20 > To: Discussion of weather data quality issues; Discussion of weather data > quality issues > Subject: Re: [wxqc] Temperature error > > FYI... Just to do a little Weather 101 (apologies in advance for the techno > stuff) > > I probably simplified the situation along the northern California coast in > my previous discussion. I was "slicing" the temperatures in your area this > morning using MADIS mesonet page and noticed stations above 1,500 ft (500 > meters) AMSL were in the 60's F (15-ish degrees C) while the stations below > 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). > > A west coast weather guy would recognize that as a classic subsidence > inversion (temperatures warm with height), but I am an east coast weather > weanie (and a little slow, I'm afraid). Checking out today's Oakland, CA, > sounding from the NWS weather balloon (aka radiosonde): > http://weather.unisys.com/upper_air/skew/skew_KOAK.html > FYI... the left-hand white line is the dew point trace, where the > temperature and dew point converge, its wet; where they diverge, its dry. > I'm assuming you are often looking down at lots of fog and low clouds from > your ranch as a result of the subsidense inversion you live in/near. > > The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" > T), you see the right hand white line (atmospheric temperature) starts > around 1000 mb (177 m or meters) at about 8 C degrees and increases with > height until it tops out at almost 20 C at about 800 meters (about 2,400 > ft). So... more than decoupling, your station (and others between 2-3,000 > ft near you) trace the warm air near the maximum temperature of the > subsidence inversion! > > Again, MADIS QCMS assumes the atmosphere cools with height which is not true > for your location in many cases. Thus, the inadvertent "flag" of your > temperature. > > > Mystery solved! > > Dave > > > > Hi Paul, > > > > MADIS QCMS uses a potential temperature comparison which corrects for > > elevation differences (dry adiabatic lapse rate) to assess station > > performance. Mesowest does a similar temperature comparison methodology, > > and also appears to show your temperatures running a little warm, after > > correcting for elevation: > > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 > > > > We have seen that valley stations run a bit "cool", and mountaintop > > stations, such as yours, run a bit "warm" in these statistical analyzes. > > We believe these warm/cool bias anomalies appear in the statistics > > because the algorithm does not account the amount of atmospheric mixing > > occurring at individual sites. Valley sites tend to "de-couple" at night > > with little atmospheric mixing and added cooling from radiational > > cooling and drainage winds. Mountaintop stations are well mixed > > throughout the night and don't typically have evening radiational > > cooling to the degree seen in valley stations. > > > > Given that you have an active aspirated temperature sensor sited in an > > open area, I think that your temperature observations are representative > > for your location. A little nit... I would lower the elevation from 15 > > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a > > big difference in performance, but that is the standard). The rain gauge > > should be similarly sited AGL to minimize potential under-catch caused > > by turbulent wind flow across the top of the rain gauge from stronger > > winds at the higher siting elevation. > > > > Dave > > CW0351 > > > > > > > > Paul Grace wrote: > > > > > The CWOP info for my station (CW0622) is showing an “error” of -4.1 > > > °F, the worst average error of -16.4. > > > > > > In looking at the neighboring stations, they are all in very different > > > areas— I am on a 640m ridge 10 miles from Pt. Arena, California, while > > > my “neighbors” are 20 – 60 miles east, in warmer valleys a few meters > > > above sea level. I would expect lower temperatures at my station. I > > > have a Davis fan-aspirated temp. sensor, that seems correct to within > > > a fraction of a degree when compared to a thermocouple. It is situated > > > in the middle of a small field, about 15ft off the ground, a couple > > > hundred feet from the nearest trees. > > > > > > How are the “correct” temperature readings > > > interpolated/extrapolated/measured? Is my station being compared to > > > these “neighbors”? > > > > > >------------------------------------------------------------------------ > > > > > >_______________________________________________ > > >wxqc mailing list > > >wxqc@lists.gladstonefamily.net > > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > > > > >The contents of this message are the responsibility of the author. > > > > > > > _______________________________________________ > > wxqc mailing list > > wxqc@lists.gladstonefamily.net > > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > > > The contents of this message are the responsibility of the author. > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. > > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. From paulgrace at lookoutranch.com Mon Nov 21 17:39:52 2005 From: paulgrace at lookoutranch.com (Paul Grace) Date: Mon Nov 21 17:40:03 2005 Subject: [wxqc] Temperature error In-Reply-To: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> Message-ID: <000301c5eeec$809de090$0a00a8c0@Dell1500> Thank you for all the help & information! -----Original Message----- From: dshelms@comcast.net [mailto:dshelms@comcast.net] Sent: Monday, November 21, 2005 14:15 To: Discussion of weather data quality issues; 'Discussion of weather data quality issues' Cc: Paul Grace Subject: RE: [wxqc] Temperature error Hi Paul, Your data are not really flagged as bad, just something that needs greater attention to see what the real deal is. Because there are very few observations from your area is a great reason for you to continue to contributing your observations. Ultimately, the local weather guys down at Monterrey and up in Eureka know from looking at the station data what is good and not so good. Again, MADIS QCMS is a first guess and not a final statement of quality. Cheers, Dave > Yes, we do look out on a river valley often full of fog, we are almost > always above it. > So you're a "slow" weather weenie? Scary. > > For a big picture question; is the fact that QCMS tags my temps/dewpoint as > bad make any difference with the use of my data? Is my data used for any > actual purpose? There are very few stations near mine, so I thought it > might be a different data point, but if it's tossed out, then should I > bother uploading it? > > -----Original Message----- > From: wxqc-bounces@lists.gladstonefamily.net > [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of > dshelms@comcast.net > Sent: Monday, November 21, 2005 11:20 > To: Discussion of weather data quality issues; Discussion of weather data > quality issues > Subject: Re: [wxqc] Temperature error > > FYI... Just to do a little Weather 101 (apologies in advance for the techno > stuff) > > I probably simplified the situation along the northern California coast in > my previous discussion. I was "slicing" the temperatures in your area this > morning using MADIS mesonet page and noticed stations above 1,500 ft (500 > meters) AMSL were in the 60's F (15-ish degrees C) while the stations below > 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). > > A west coast weather guy would recognize that as a classic subsidence > inversion (temperatures warm with height), but I am an east coast weather > weanie (and a little slow, I'm afraid). Checking out today's Oakland, CA, > sounding from the NWS weather balloon (aka radiosonde): > http://weather.unisys.com/upper_air/skew/skew_KOAK.html > FYI... the left-hand white line is the dew point trace, where the > temperature and dew point converge, its wet; where they diverge, its dry. > I'm assuming you are often looking down at lots of fog and low clouds from > your ranch as a result of the subsidense inversion you live in/near. > > The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" > T), you see the right hand white line (atmospheric temperature) starts > around 1000 mb (177 m or meters) at about 8 C degrees and increases with > height until it tops out at almost 20 C at about 800 meters (about 2,400 > ft). So... more than decoupling, your station (and others between 2-3,000 > ft near you) trace the warm air near the maximum temperature of the > subsidence inversion! > > Again, MADIS QCMS assumes the atmosphere cools with height which is not true > for your location in many cases. Thus, the inadvertent "flag" of your > temperature. > > > Mystery solved! > > Dave > > > > Hi Paul, > > > > MADIS QCMS uses a potential temperature comparison which corrects for > > elevation differences (dry adiabatic lapse rate) to assess station > > performance. Mesowest does a similar temperature comparison methodology, > > and also appears to show your temperatures running a little warm, after > > correcting for elevation: > > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 > > > > We have seen that valley stations run a bit "cool", and mountaintop > > stations, such as yours, run a bit "warm" in these statistical analyzes. > > We believe these warm/cool bias anomalies appear in the statistics > > because the algorithm does not account the amount of atmospheric mixing > > occurring at individual sites. Valley sites tend to "de-couple" at night > > with little atmospheric mixing and added cooling from radiational > > cooling and drainage winds. Mountaintop stations are well mixed > > throughout the night and don't typically have evening radiational > > cooling to the degree seen in valley stations. > > > > Given that you have an active aspirated temperature sensor sited in an > > open area, I think that your temperature observations are representative > > for your location. A little nit... I would lower the elevation from 15 > > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a > > big difference in performance, but that is the standard). The rain gauge > > should be similarly sited AGL to minimize potential under-catch caused > > by turbulent wind flow across the top of the rain gauge from stronger > > winds at the higher siting elevation. > > > > Dave > > CW0351 > > > > > > > > Paul Grace wrote: > > > > > The CWOP info for my station (CW0622) is showing an ?error? of -4.1 > > > ?F, the worst average error of -16.4. > > > > > > In looking at the neighboring stations, they are all in very different > > > areas? I am on a 640m ridge 10 miles from Pt. Arena, California, while > > > my ?neighbors? are 20 ? 60 miles east, in warmer valleys a few meters > > > above sea level. I would expect lower temperatures at my station. I > > > have a Davis fan-aspirated temp. sensor, that seems correct to within > > > a fraction of a degree when compared to a thermocouple. It is situated > > > in the middle of a small field, about 15ft off the ground, a couple > > > hundred feet from the nearest trees. > > > > > > How are the ?correct? temperature readings > > > interpolated/extrapolated/measured? Is my station being compared to > > > these ?neighbors?? > > > > > >------------------------------------------------------------------------ > > > > > >_______________________________________________ > > >wxqc mailing list > > >wxqc@lists.gladstonefamily.net > > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > > > > >The contents of this message are the responsibility of the author. > > > > > > > _______________________________________________ > > wxqc mailing list > > wxqc@lists.gladstonefamily.net > > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > > > The contents of this message are the responsibility of the author. > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. > > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. From jrjones26 at direcway.com Mon Nov 21 23:23:14 2005 From: jrjones26 at direcway.com (Jack Jones) Date: Mon Nov 21 23:23:18 2005 Subject: [wxqc] Temperature error References: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> Message-ID: <002501c5ef1c$76e7e980$6401a8c0@intel3200> Hi Dave, These discussions and others that have occurred on this forum show just how screwy the "global warming" folks are. Your models cannot accurately determine temperatures in a 20 or 30 mile radius due to terrain differences or other anomolies, yet, our "global warming" enthusiasts claim to be able to predict the temperatures over the entire planet. Those of you who work with the models seem to get upset when someone reports data that is dramatically different than the forecast. The first choice is to cast aspersions on the quality/accuracy of the measuring device but ever on the quality of the algorithms of the model. Some of you have the attitude of many scientists I know, which is " don't confuse me with the facts". I have worked with several models for chemical reactions, water temperature predictions, algae growth, oxygen depletion in lakes and others and basically they are only as good as the few data points available & the mathematical formulae used to try to simulate the data. However, that doesn't deter the true scientist from questioning any data that deviates from the model. Jack Jones ----- Original Message ----- From: To: "Discussion of weather data quality issues" ; "'Discussion of weather data quality issues'" Sent: Monday, November 21, 2005 4:15 PM Subject: RE: [wxqc] Temperature error > Hi Paul, > > Your data are not really flagged as bad, just something that needs greater > attention to see what the real deal is. Because there are very few > observations from your area is a great reason for you to continue to > contributing your observations. Ultimately, the local weather guys down > at Monterrey and up in Eureka know from looking at the station data what > is good and not so good. > > Again, MADIS QCMS is a first guess and not a final statement of quality. > > Cheers, > > Dave > > >> Yes, we do look out on a river valley often full of fog, we are almost >> always above it. >> So you're a "slow" weather weenie? Scary. >> >> For a big picture question; is the fact that QCMS tags my temps/dewpoint >> as >> bad make any difference with the use of my data? Is my data used for any >> actual purpose? There are very few stations near mine, so I thought it >> might be a different data point, but if it's tossed out, then should I >> bother uploading it? >> >> -----Original Message----- >> From: wxqc-bounces@lists.gladstonefamily.net >> [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of >> dshelms@comcast.net >> Sent: Monday, November 21, 2005 11:20 >> To: Discussion of weather data quality issues; Discussion of weather data >> quality issues >> Subject: Re: [wxqc] Temperature error >> >> FYI... Just to do a little Weather 101 (apologies in advance for the >> techno >> stuff) >> >> I probably simplified the situation along the northern California coast >> in >> my previous discussion. I was "slicing" the temperatures in your area >> this >> morning using MADIS mesonet page and noticed stations above 1,500 ft (500 >> meters) AMSL were in the 60's F (15-ish degrees C) while the stations >> below >> 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). >> >> A west coast weather guy would recognize that as a classic subsidence >> inversion (temperatures warm with height), but I am an east coast weather >> weanie (and a little slow, I'm afraid). Checking out today's Oakland, >> CA, >> sounding from the NWS weather balloon (aka radiosonde): >> http://weather.unisys.com/upper_air/skew/skew_KOAK.html >> FYI... the left-hand white line is the dew point trace, where the >> temperature and dew point converge, its wet; where they diverge, its dry. >> I'm assuming you are often looking down at lots of fog and low clouds >> from >> your ranch as a result of the subsidense inversion you live in/near. >> >> The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" >> T), you see the right hand white line (atmospheric temperature) starts >> around 1000 mb (177 m or meters) at about 8 C degrees and increases with >> height until it tops out at almost 20 C at about 800 meters (about 2,400 >> ft). So... more than decoupling, your station (and others between >> 2-3,000 >> ft near you) trace the warm air near the maximum temperature of the >> subsidence inversion! >> >> Again, MADIS QCMS assumes the atmosphere cools with height which is not >> true >> for your location in many cases. Thus, the inadvertent "flag" of your >> temperature. >> >> >> Mystery solved! >> >> Dave >> >> >> > Hi Paul, >> > >> > MADIS QCMS uses a potential temperature comparison which corrects for >> > elevation differences (dry adiabatic lapse rate) to assess station >> > performance. Mesowest does a similar temperature comparison >> > methodology, >> > and also appears to show your temperatures running a little warm, after >> > correcting for elevation: >> > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 >> > >> > We have seen that valley stations run a bit "cool", and mountaintop >> > stations, such as yours, run a bit "warm" in these statistical >> > analyzes. >> > We believe these warm/cool bias anomalies appear in the statistics >> > because the algorithm does not account the amount of atmospheric mixing >> > occurring at individual sites. Valley sites tend to "de-couple" at >> > night >> > with little atmospheric mixing and added cooling from radiational >> > cooling and drainage winds. Mountaintop stations are well mixed >> > throughout the night and don't typically have evening radiational >> > cooling to the degree seen in valley stations. >> > >> > Given that you have an active aspirated temperature sensor sited in an >> > open area, I think that your temperature observations are >> > representative >> > for your location. A little nit... I would lower the elevation from 15 >> > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a >> > big difference in performance, but that is the standard). The rain >> > gauge >> > should be similarly sited AGL to minimize potential under-catch caused >> > by turbulent wind flow across the top of the rain gauge from stronger >> > winds at the higher siting elevation. >> > >> > Dave >> > CW0351 >> > >> > >> > >> > Paul Grace wrote: >> > >> > > The CWOP info for my station (CW0622) is showing an "error" of -4.1 >> > > ?F, the worst average error of -16.4. >> > > >> > > In looking at the neighboring stations, they are all in very >> > > different >> > > areas- I am on a 640m ridge 10 miles from Pt. Arena, California, >> > > while >> > > my "neighbors" are 20 - 60 miles east, in warmer valleys a few meters >> > > above sea level. I would expect lower temperatures at my station. I >> > > have a Davis fan-aspirated temp. sensor, that seems correct to within >> > > a fraction of a degree when compared to a thermocouple. It is >> > > situated >> > > in the middle of a small field, about 15ft off the ground, a couple >> > > hundred feet from the nearest trees. >> > > >> > > How are the "correct" temperature readings >> > > interpolated/extrapolated/measured? Is my station being compared to >> > > these "neighbors"? >> > > >> > >------------------------------------------------------------------------ >> > > >> > >_______________________________________________ >> > >wxqc mailing list >> > >wxqc@lists.gladstonefamily.net >> > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > > >> > >The contents of this message are the responsibility of the author. >> > > >> > >> > _______________________________________________ >> > wxqc mailing list >> > wxqc@lists.gladstonefamily.net >> > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > >> > The contents of this message are the responsibility of the author. >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. >> >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. From paulgrace at lookoutranch.com Mon Nov 21 23:37:30 2005 From: paulgrace at lookoutranch.com (Paul Grace) Date: Mon Nov 21 23:37:46 2005 Subject: [wxqc] Temperature error In-Reply-To: <002501c5ef1c$76e7e980$6401a8c0@intel3200> Message-ID: <000001c5ef1e$79c33220$0a00a8c0@Dell1500> I don't know if this is a good forum to be calling people that believe humans are making global warming worse "screwy". -----Original Message----- From: wxqc-bounces@lists.gladstonefamily.net [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of Jack Jones Sent: Monday, November 21, 2005 20:23 To: Discussion of weather data quality issues Subject: Re: [wxqc] Temperature error Hi Dave, These discussions and others that have occurred on this forum show just how screwy the "global warming" folks are. Your models cannot accurately determine temperatures in a 20 or 30 mile radius due to terrain differences or other anomolies, yet, our "global warming" enthusiasts claim to be able to predict the temperatures over the entire planet. Those of you who work with the models seem to get upset when someone reports data that is dramatically different than the forecast. The first choice is to cast aspersions on the quality/accuracy of the measuring device but ever on the quality of the algorithms of the model. Some of you have the attitude of many scientists I know, which is " don't confuse me with the facts". I have worked with several models for chemical reactions, water temperature predictions, algae growth, oxygen depletion in lakes and others and basically they are only as good as the few data points available & the mathematical formulae used to try to simulate the data. However, that doesn't deter the true scientist from questioning any data that deviates from the model. Jack Jones ----- Original Message ----- From: To: "Discussion of weather data quality issues" ; "'Discussion of weather data quality issues'" Sent: Monday, November 21, 2005 4:15 PM Subject: RE: [wxqc] Temperature error > Hi Paul, > > Your data are not really flagged as bad, just something that needs greater > attention to see what the real deal is. Because there are very few > observations from your area is a great reason for you to continue to > contributing your observations. Ultimately, the local weather guys down > at Monterrey and up in Eureka know from looking at the station data what > is good and not so good. > > Again, MADIS QCMS is a first guess and not a final statement of quality. > > Cheers, > > Dave > > >> Yes, we do look out on a river valley often full of fog, we are almost >> always above it. >> So you're a "slow" weather weenie? Scary. >> >> For a big picture question; is the fact that QCMS tags my temps/dewpoint >> as >> bad make any difference with the use of my data? Is my data used for any >> actual purpose? There are very few stations near mine, so I thought it >> might be a different data point, but if it's tossed out, then should I >> bother uploading it? >> >> -----Original Message----- >> From: wxqc-bounces@lists.gladstonefamily.net >> [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of >> dshelms@comcast.net >> Sent: Monday, November 21, 2005 11:20 >> To: Discussion of weather data quality issues; Discussion of weather data >> quality issues >> Subject: Re: [wxqc] Temperature error >> >> FYI... Just to do a little Weather 101 (apologies in advance for the >> techno >> stuff) >> >> I probably simplified the situation along the northern California coast >> in >> my previous discussion. I was "slicing" the temperatures in your area >> this >> morning using MADIS mesonet page and noticed stations above 1,500 ft (500 >> meters) AMSL were in the 60's F (15-ish degrees C) while the stations >> below >> 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). >> >> A west coast weather guy would recognize that as a classic subsidence >> inversion (temperatures warm with height), but I am an east coast weather >> weanie (and a little slow, I'm afraid). Checking out today's Oakland, >> CA, >> sounding from the NWS weather balloon (aka radiosonde): >> http://weather.unisys.com/upper_air/skew/skew_KOAK.html >> FYI... the left-hand white line is the dew point trace, where the >> temperature and dew point converge, its wet; where they diverge, its dry. >> I'm assuming you are often looking down at lots of fog and low clouds >> from >> your ranch as a result of the subsidense inversion you live in/near. >> >> The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" >> T), you see the right hand white line (atmospheric temperature) starts >> around 1000 mb (177 m or meters) at about 8 C degrees and increases with >> height until it tops out at almost 20 C at about 800 meters (about 2,400 >> ft). So... more than decoupling, your station (and others between >> 2-3,000 >> ft near you) trace the warm air near the maximum temperature of the >> subsidence inversion! >> >> Again, MADIS QCMS assumes the atmosphere cools with height which is not >> true >> for your location in many cases. Thus, the inadvertent "flag" of your >> temperature. >> >> >> Mystery solved! >> >> Dave >> >> >> > Hi Paul, >> > >> > MADIS QCMS uses a potential temperature comparison which corrects for >> > elevation differences (dry adiabatic lapse rate) to assess station >> > performance. Mesowest does a similar temperature comparison >> > methodology, >> > and also appears to show your temperatures running a little warm, after >> > correcting for elevation: >> > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 >> > >> > We have seen that valley stations run a bit "cool", and mountaintop >> > stations, such as yours, run a bit "warm" in these statistical >> > analyzes. >> > We believe these warm/cool bias anomalies appear in the statistics >> > because the algorithm does not account the amount of atmospheric mixing >> > occurring at individual sites. Valley sites tend to "de-couple" at >> > night >> > with little atmospheric mixing and added cooling from radiational >> > cooling and drainage winds. Mountaintop stations are well mixed >> > throughout the night and don't typically have evening radiational >> > cooling to the degree seen in valley stations. >> > >> > Given that you have an active aspirated temperature sensor sited in an >> > open area, I think that your temperature observations are >> > representative >> > for your location. A little nit... I would lower the elevation from 15 >> > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a >> > big difference in performance, but that is the standard). The rain >> > gauge >> > should be similarly sited AGL to minimize potential under-catch caused >> > by turbulent wind flow across the top of the rain gauge from stronger >> > winds at the higher siting elevation. >> > >> > Dave >> > CW0351 >> > >> > >> > >> > Paul Grace wrote: >> > >> > > The CWOP info for my station (CW0622) is showing an "error" of -4.1 >> > > ?F, the worst average error of -16.4. >> > > >> > > In looking at the neighboring stations, they are all in very >> > > different >> > > areas- I am on a 640m ridge 10 miles from Pt. Arena, California, >> > > while >> > > my "neighbors" are 20 - 60 miles east, in warmer valleys a few meters >> > > above sea level. I would expect lower temperatures at my station. I >> > > have a Davis fan-aspirated temp. sensor, that seems correct to within >> > > a fraction of a degree when compared to a thermocouple. It is >> > > situated >> > > in the middle of a small field, about 15ft off the ground, a couple >> > > hundred feet from the nearest trees. >> > > >> > > How are the "correct" temperature readings >> > > interpolated/extrapolated/measured? Is my station being compared to >> > > these "neighbors"? >> > > >> > >------------------------------------------------------------------------ >> > > >> > >_______________________________________________ >> > >wxqc mailing list >> > >wxqc@lists.gladstonefamily.net >> > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > > >> > >The contents of this message are the responsibility of the author. >> > > >> > >> > _______________________________________________ >> > wxqc mailing list >> > wxqc@lists.gladstonefamily.net >> > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > >> > The contents of this message are the responsibility of the author. >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. >> >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. _______________________________________________ wxqc mailing list wxqc@lists.gladstonefamily.net http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc The contents of this message are the responsibility of the author. From dshelms at comcast.net Mon Nov 21 23:40:07 2005 From: dshelms at comcast.net (Dave Helms) Date: Mon Nov 21 23:40:11 2005 Subject: [wxqc] Temperature error In-Reply-To: <2AA5D5B18E1F4849A2B998B43BAA2BA70FC6FA@fjuckby.coursey.local> References: <2AA5D5B18E1F4849A2B998B43BAA2BA70FC6FA@fjuckby.coursey.local> Message-ID: <4382A127.5000807@comcast.net> Hi David, Well, your station does have three green checks so your data is not too far off what MADIS QCMS expects: http://weather.gladstonefamily.net/site/AR987 The only instance where your data appears to diverge from the MADIS QCMS analysis is at minimum temperature just before sunrise when your station is cooler than the station in your vicinity. As a valley station, this is not too unusual as drainage winds will influence your minimum temperatures (this is the opposite of Paul's mountaintop situation). The dew point analysis also seems to have a mild case of the jaggies which is caused by a station with a too-dry dew point (I suspect its the Rose Peak RAWS station, RSPC1). Hope this helps, Dave CW0351 * *David Coursey wrote: >I have a similar problem here at AR987 N5FDL in Tracy, Calif. My temps seem way off sometimes, though when I check them with another thermometer they always check out. I am in the Central Valley (by a few mines, but across the Altamont Range) some of the stations that I think I am being compared to are back toward the Bay Area and on the other side of the hills. Could someone take a look at my temps and tell me if they look OK? I am concerned that I am sending bad data, even though I am pretty sure I am not. > >My station is a Davis Vantage Pro II mounted 5-6 feet off a concrete tile roof at about 32 feet AGL. > >Thanks, > >David, N5FDL > >-----Original Message----- >From: wxqc-bounces@lists.gladstonefamily.net [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of Dave Helms >Sent: Sunday, November 20, 2005 9:46 PM >To: Discussion of weather data quality issues >Subject: Re: [wxqc] Temperature error > >Hi Paul, > >MADIS QCMS uses a potential temperature comparison which corrects for elevation differences (dry adiabatic lapse rate) to assess station performance. Mesowest does a similar temperature comparison methodology, and also appears to show your temperatures running a little warm, after correcting for elevation: >http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 > >We have seen that valley stations run a bit "cool", and mountaintop stations, such as yours, run a bit "warm" in these statistical analyzes. >We believe these warm/cool bias anomalies appear in the statistics because the algorithm does not account the amount of atmospheric mixing occurring at individual sites. Valley sites tend to "de-couple" at night with little atmospheric mixing and added cooling from radiational cooling and drainage winds. Mountaintop stations are well mixed throughout the night and don't typically have evening radiational cooling to the degree seen in valley stations. > >Given that you have an active aspirated temperature sensor sited in an open area, I think that your temperature observations are representative for your location. A little nit... I would lower the elevation from 15 ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a big difference in performance, but that is the standard). The rain gauge should be similarly sited AGL to minimize potential under-catch caused by turbulent wind flow across the top of the rain gauge from stronger winds at the higher siting elevation. > >Dave >CW0351 > > > >Paul Grace wrote: > > > >>The CWOP info for my station (CW0622) is showing an "error" of -4.1 >>?F, the worst average error of -16.4. >> >>In looking at the neighboring stations, they are all in very different >>areas- I am on a 640m ridge 10 miles from Pt. Arena, California, while >>my "neighbors" are 20 - 60 miles east, in warmer valleys a few meters >>above sea level. I would expect lower temperatures at my station. I >>have a Davis fan-aspirated temp. sensor, that seems correct to within >>a fraction of a degree when compared to a thermocouple. It is situated >>in the middle of a small field, about 15ft off the ground, a couple >>hundred feet from the nearest trees. >> >>How are the "correct" temperature readings >>interpolated/extrapolated/measured? Is my station being compared to >>these "neighbors"? >> >>----------------------------------------------------------------------- >>- >> >>_______________________________________________ >>wxqc mailing list >>wxqc@lists.gladstonefamily.net >>http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >>The contents of this message are the responsibility of the author. >> >> >> > >_______________________________________________ >wxqc mailing list >wxqc@lists.gladstonefamily.net >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > >The contents of this message are the responsibility of the author. >e responsibility of the author. >_______________________________________________ >wxqc mailing list >wxqc@lists.gladstonefamily.net >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > >The contents of this message are the responsibility of the author. > > > > From jrjones26 at direcway.com Mon Nov 21 23:43:03 2005 From: jrjones26 at direcway.com (Jack Jones) Date: Mon Nov 21 23:42:59 2005 Subject: [wxqc] Temperature error References: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> Message-ID: <003f01c5ef1f$3b025380$6401a8c0@intel3200> Hi dave, Another point. From some of the previous comments, I assume yor models can predict temperatue and wind profiles by elevation above grade, i.e. 5,10,15,20,30,50 ft., since you seem to claim that an elevation higher than 5 ft. could explain the difference between observed and predicted values. >From this comment & a previous one, you might assume that I don't like your models. That isn't true, what I don't like is the cavalier attitude toward data that differs from your models. There are measuring devices that can be very inaccurate and to identify and highlight that is good. Yes, one can scrape ice off his windshield at temps of 34 to 35 deg F when a device says it's 32, but if your model says it's 40 deg F, your model is wrong. I thought the whole concept of this data collection was to help the model, but, it seems the data received is mostly ignored if it doesn't agree with "the model". Jack Jones ----- Original Message ----- From: To: "Discussion of weather data quality issues" ; "'Discussion of weather data quality issues'" Sent: Monday, November 21, 2005 4:15 PM Subject: RE: [wxqc] Temperature error > Hi Paul, > > Your data are not really flagged as bad, just something that needs greater > attention to see what the real deal is. Because there are very few > observations from your area is a great reason for you to continue to > contributing your observations. Ultimately, the local weather guys down > at Monterrey and up in Eureka know from looking at the station data what > is good and not so good. > > Again, MADIS QCMS is a first guess and not a final statement of quality. > > Cheers, > > Dave > > >> Yes, we do look out on a river valley often full of fog, we are almost >> always above it. >> So you're a "slow" weather weenie? Scary. >> >> For a big picture question; is the fact that QCMS tags my temps/dewpoint >> as >> bad make any difference with the use of my data? Is my data used for any >> actual purpose? There are very few stations near mine, so I thought it >> might be a different data point, but if it's tossed out, then should I >> bother uploading it? >> >> -----Original Message----- >> From: wxqc-bounces@lists.gladstonefamily.net >> [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of >> dshelms@comcast.net >> Sent: Monday, November 21, 2005 11:20 >> To: Discussion of weather data quality issues; Discussion of weather data >> quality issues >> Subject: Re: [wxqc] Temperature error >> >> FYI... Just to do a little Weather 101 (apologies in advance for the >> techno >> stuff) >> >> I probably simplified the situation along the northern California coast >> in >> my previous discussion. I was "slicing" the temperatures in your area >> this >> morning using MADIS mesonet page and noticed stations above 1,500 ft (500 >> meters) AMSL were in the 60's F (15-ish degrees C) while the stations >> below >> 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). >> >> A west coast weather guy would recognize that as a classic subsidence >> inversion (temperatures warm with height), but I am an east coast weather >> weanie (and a little slow, I'm afraid). Checking out today's Oakland, >> CA, >> sounding from the NWS weather balloon (aka radiosonde): >> http://weather.unisys.com/upper_air/skew/skew_KOAK.html >> FYI... the left-hand white line is the dew point trace, where the >> temperature and dew point converge, its wet; where they diverge, its dry. >> I'm assuming you are often looking down at lots of fog and low clouds >> from >> your ranch as a result of the subsidense inversion you live in/near. >> >> The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" >> T), you see the right hand white line (atmospheric temperature) starts >> around 1000 mb (177 m or meters) at about 8 C degrees and increases with >> height until it tops out at almost 20 C at about 800 meters (about 2,400 >> ft). So... more than decoupling, your station (and others between >> 2-3,000 >> ft near you) trace the warm air near the maximum temperature of the >> subsidence inversion! >> >> Again, MADIS QCMS assumes the atmosphere cools with height which is not >> true >> for your location in many cases. Thus, the inadvertent "flag" of your >> temperature. >> >> >> Mystery solved! >> >> Dave >> >> >> > Hi Paul, >> > >> > MADIS QCMS uses a potential temperature comparison which corrects for >> > elevation differences (dry adiabatic lapse rate) to assess station >> > performance. Mesowest does a similar temperature comparison >> > methodology, >> > and also appears to show your temperatures running a little warm, after >> > correcting for elevation: >> > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 >> > >> > We have seen that valley stations run a bit "cool", and mountaintop >> > stations, such as yours, run a bit "warm" in these statistical >> > analyzes. >> > We believe these warm/cool bias anomalies appear in the statistics >> > because the algorithm does not account the amount of atmospheric mixing >> > occurring at individual sites. Valley sites tend to "de-couple" at >> > night >> > with little atmospheric mixing and added cooling from radiational >> > cooling and drainage winds. Mountaintop stations are well mixed >> > throughout the night and don't typically have evening radiational >> > cooling to the degree seen in valley stations. >> > >> > Given that you have an active aspirated temperature sensor sited in an >> > open area, I think that your temperature observations are >> > representative >> > for your location. A little nit... I would lower the elevation from 15 >> > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a >> > big difference in performance, but that is the standard). The rain >> > gauge >> > should be similarly sited AGL to minimize potential under-catch caused >> > by turbulent wind flow across the top of the rain gauge from stronger >> > winds at the higher siting elevation. >> > >> > Dave >> > CW0351 >> > >> > >> > >> > Paul Grace wrote: >> > >> > > The CWOP info for my station (CW0622) is showing an "error" of -4.1 >> > > ?F, the worst average error of -16.4. >> > > >> > > In looking at the neighboring stations, they are all in very >> > > different >> > > areas- I am on a 640m ridge 10 miles from Pt. Arena, California, >> > > while >> > > my "neighbors" are 20 - 60 miles east, in warmer valleys a few meters >> > > above sea level. I would expect lower temperatures at my station. I >> > > have a Davis fan-aspirated temp. sensor, that seems correct to within >> > > a fraction of a degree when compared to a thermocouple. It is >> > > situated >> > > in the middle of a small field, about 15ft off the ground, a couple >> > > hundred feet from the nearest trees. >> > > >> > > How are the "correct" temperature readings >> > > interpolated/extrapolated/measured? Is my station being compared to >> > > these "neighbors"? >> > > >> > >------------------------------------------------------------------------ >> > > >> > >_______________________________________________ >> > >wxqc mailing list >> > >wxqc@lists.gladstonefamily.net >> > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > > >> > >The contents of this message are the responsibility of the author. >> > > >> > >> > _______________________________________________ >> > wxqc mailing list >> > wxqc@lists.gladstonefamily.net >> > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > >> > The contents of this message are the responsibility of the author. >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. >> >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. From jrjones26 at direcway.com Mon Nov 21 23:50:26 2005 From: jrjones26 at direcway.com (Jack Jones) Date: Mon Nov 21 23:50:25 2005 Subject: [wxqc] Temperature error References: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> Message-ID: <004201c5ef20$437ee9f0$6401a8c0@intel3200> Hi Dave, You weather guys are a "trip". If I had been as wrong as often as weather forecasters in my previous job, I would have been unemployed for the last 30 years. However, weather forecasters hve the audacity to present a model to forecast temperatures & other weather data, then "cry foul" when someone presents them with data that is different from the forecast. How many times have I carried an umbrella on sunny days just because a weather forecaster said it will rain. Jack Jones ----- Original Message ----- From: To: "Discussion of weather data quality issues" ; "'Discussion of weather data quality issues'" Sent: Monday, November 21, 2005 4:15 PM Subject: RE: [wxqc] Temperature error > Hi Paul, > > Your data are not really flagged as bad, just something that needs greater > attention to see what the real deal is. Because there are very few > observations from your area is a great reason for you to continue to > contributing your observations. Ultimately, the local weather guys down > at Monterrey and up in Eureka know from looking at the station data what > is good and not so good. > > Again, MADIS QCMS is a first guess and not a final statement of quality. > > Cheers, > > Dave > > >> Yes, we do look out on a river valley often full of fog, we are almost >> always above it. >> So you're a "slow" weather weenie? Scary. >> >> For a big picture question; is the fact that QCMS tags my temps/dewpoint >> as >> bad make any difference with the use of my data? Is my data used for any >> actual purpose? There are very few stations near mine, so I thought it >> might be a different data point, but if it's tossed out, then should I >> bother uploading it? >> >> -----Original Message----- >> From: wxqc-bounces@lists.gladstonefamily.net >> [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of >> dshelms@comcast.net >> Sent: Monday, November 21, 2005 11:20 >> To: Discussion of weather data quality issues; Discussion of weather data >> quality issues >> Subject: Re: [wxqc] Temperature error >> >> FYI... Just to do a little Weather 101 (apologies in advance for the >> techno >> stuff) >> >> I probably simplified the situation along the northern California coast >> in >> my previous discussion. I was "slicing" the temperatures in your area >> this >> morning using MADIS mesonet page and noticed stations above 1,500 ft (500 >> meters) AMSL were in the 60's F (15-ish degrees C) while the stations >> below >> 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). >> >> A west coast weather guy would recognize that as a classic subsidence >> inversion (temperatures warm with height), but I am an east coast weather >> weanie (and a little slow, I'm afraid). Checking out today's Oakland, >> CA, >> sounding from the NWS weather balloon (aka radiosonde): >> http://weather.unisys.com/upper_air/skew/skew_KOAK.html >> FYI... the left-hand white line is the dew point trace, where the >> temperature and dew point converge, its wet; where they diverge, its dry. >> I'm assuming you are often looking down at lots of fog and low clouds >> from >> your ranch as a result of the subsidense inversion you live in/near. >> >> The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" >> T), you see the right hand white line (atmospheric temperature) starts >> around 1000 mb (177 m or meters) at about 8 C degrees and increases with >> height until it tops out at almost 20 C at about 800 meters (about 2,400 >> ft). So... more than decoupling, your station (and others between >> 2-3,000 >> ft near you) trace the warm air near the maximum temperature of the >> subsidence inversion! >> >> Again, MADIS QCMS assumes the atmosphere cools with height which is not >> true >> for your location in many cases. Thus, the inadvertent "flag" of your >> temperature. >> >> >> Mystery solved! >> >> Dave >> >> >> > Hi Paul, >> > >> > MADIS QCMS uses a potential temperature comparison which corrects for >> > elevation differences (dry adiabatic lapse rate) to assess station >> > performance. Mesowest does a similar temperature comparison >> > methodology, >> > and also appears to show your temperatures running a little warm, after >> > correcting for elevation: >> > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 >> > >> > We have seen that valley stations run a bit "cool", and mountaintop >> > stations, such as yours, run a bit "warm" in these statistical >> > analyzes. >> > We believe these warm/cool bias anomalies appear in the statistics >> > because the algorithm does not account the amount of atmospheric mixing >> > occurring at individual sites. Valley sites tend to "de-couple" at >> > night >> > with little atmospheric mixing and added cooling from radiational >> > cooling and drainage winds. Mountaintop stations are well mixed >> > throughout the night and don't typically have evening radiational >> > cooling to the degree seen in valley stations. >> > >> > Given that you have an active aspirated temperature sensor sited in an >> > open area, I think that your temperature observations are >> > representative >> > for your location. A little nit... I would lower the elevation from 15 >> > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a >> > big difference in performance, but that is the standard). The rain >> > gauge >> > should be similarly sited AGL to minimize potential under-catch caused >> > by turbulent wind flow across the top of the rain gauge from stronger >> > winds at the higher siting elevation. >> > >> > Dave >> > CW0351 >> > >> > >> > >> > Paul Grace wrote: >> > >> > > The CWOP info for my station (CW0622) is showing an "error" of -4.1 >> > > ?F, the worst average error of -16.4. >> > > >> > > In looking at the neighboring stations, they are all in very >> > > different >> > > areas- I am on a 640m ridge 10 miles from Pt. Arena, California, >> > > while >> > > my "neighbors" are 20 - 60 miles east, in warmer valleys a few meters >> > > above sea level. I would expect lower temperatures at my station. I >> > > have a Davis fan-aspirated temp. sensor, that seems correct to within >> > > a fraction of a degree when compared to a thermocouple. It is >> > > situated >> > > in the middle of a small field, about 15ft off the ground, a couple >> > > hundred feet from the nearest trees. >> > > >> > > How are the "correct" temperature readings >> > > interpolated/extrapolated/measured? Is my station being compared to >> > > these "neighbors"? >> > > >> > >------------------------------------------------------------------------ >> > > >> > >_______________________________________________ >> > >wxqc mailing list >> > >wxqc@lists.gladstonefamily.net >> > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > > >> > >The contents of this message are the responsibility of the author. >> > > >> > >> > _______________________________________________ >> > wxqc mailing list >> > wxqc@lists.gladstonefamily.net >> > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > >> > The contents of this message are the responsibility of the author. >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. >> >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. From jrjones26 at direcway.com Tue Nov 22 00:05:34 2005 From: jrjones26 at direcway.com (Jack Jones) Date: Tue Nov 22 00:05:33 2005 Subject: [wxqc] Temperature error References: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> Message-ID: <004501c5ef22$608e9250$6401a8c0@intel3200> Hi Dave, By now you may have deduced that I am a little tired of the sanctimonious responses by some of the weather "professionals" on this forum. I doubt that "the local weather guys down at Monterrey and up in Eureka" know which way is up, much less, what is good or bad info. Most of them read what they get from the computer and try to deliver it on air as if they had a clue, which mostly they don't. I recall Joe Bastardi on Accuweather having the audacity to predict hurricane Rita's path and landfall 24 hours ahead of landfall, saying he couuld predict within 40 miles. He missed it by 60 miles. So please give us a break. It's okay to explain what the model is intended to do, just don't believe what it forecasts is gospel and cut those dissenters with multiple data points some slack. Jack Jones Jack Jones ----- Original Message ----- From: To: "Discussion of weather data quality issues" ; "'Discussion of weather data quality issues'" Sent: Monday, November 21, 2005 4:15 PM Subject: RE: [wxqc] Temperature error > Hi Paul, > > Your data are not really flagged as bad, just something that needs greater > attention to see what the real deal is. Because there are very few > observations from your area is a great reason for you to continue to > contributing your observations. Ultimately, the local weather guys down > at Monterrey and up in Eureka know from looking at the station data what > is good and not so good. > > Again, MADIS QCMS is a first guess and not a final statement of quality. > > Cheers, > > Dave > > >> Yes, we do look out on a river valley often full of fog, we are almost >> always above it. >> So you're a "slow" weather weenie? Scary. >> >> For a big picture question; is the fact that QCMS tags my temps/dewpoint >> as >> bad make any difference with the use of my data? Is my data used for any >> actual purpose? There are very few stations near mine, so I thought it >> might be a different data point, but if it's tossed out, then should I >> bother uploading it? >> >> -----Original Message----- >> From: wxqc-bounces@lists.gladstonefamily.net >> [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of >> dshelms@comcast.net >> Sent: Monday, November 21, 2005 11:20 >> To: Discussion of weather data quality issues; Discussion of weather data >> quality issues >> Subject: Re: [wxqc] Temperature error >> >> FYI... Just to do a little Weather 101 (apologies in advance for the >> techno >> stuff) >> >> I probably simplified the situation along the northern California coast >> in >> my previous discussion. I was "slicing" the temperatures in your area >> this >> morning using MADIS mesonet page and noticed stations above 1,500 ft (500 >> meters) AMSL were in the 60's F (15-ish degrees C) while the stations >> below >> 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). >> >> A west coast weather guy would recognize that as a classic subsidence >> inversion (temperatures warm with height), but I am an east coast weather >> weanie (and a little slow, I'm afraid). Checking out today's Oakland, >> CA, >> sounding from the NWS weather balloon (aka radiosonde): >> http://weather.unisys.com/upper_air/skew/skew_KOAK.html >> FYI... the left-hand white line is the dew point trace, where the >> temperature and dew point converge, its wet; where they diverge, its dry. >> I'm assuming you are often looking down at lots of fog and low clouds >> from >> your ranch as a result of the subsidense inversion you live in/near. >> >> The diagnal solid blue lines are the temperature isotherms (e.g. "skewed" >> T), you see the right hand white line (atmospheric temperature) starts >> around 1000 mb (177 m or meters) at about 8 C degrees and increases with >> height until it tops out at almost 20 C at about 800 meters (about 2,400 >> ft). So... more than decoupling, your station (and others between >> 2-3,000 >> ft near you) trace the warm air near the maximum temperature of the >> subsidence inversion! >> >> Again, MADIS QCMS assumes the atmosphere cools with height which is not >> true >> for your location in many cases. Thus, the inadvertent "flag" of your >> temperature. >> >> >> Mystery solved! >> >> Dave >> >> >> > Hi Paul, >> > >> > MADIS QCMS uses a potential temperature comparison which corrects for >> > elevation differences (dry adiabatic lapse rate) to assess station >> > performance. Mesowest does a similar temperature comparison >> > methodology, >> > and also appears to show your temperatures running a little warm, after >> > correcting for elevation: >> > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 >> > >> > We have seen that valley stations run a bit "cool", and mountaintop >> > stations, such as yours, run a bit "warm" in these statistical >> > analyzes. >> > We believe these warm/cool bias anomalies appear in the statistics >> > because the algorithm does not account the amount of atmospheric mixing >> > occurring at individual sites. Valley sites tend to "de-couple" at >> > night >> > with little atmospheric mixing and added cooling from radiational >> > cooling and drainage winds. Mountaintop stations are well mixed >> > throughout the night and don't typically have evening radiational >> > cooling to the degree seen in valley stations. >> > >> > Given that you have an active aspirated temperature sensor sited in an >> > open area, I think that your temperature observations are >> > representative >> > for your location. A little nit... I would lower the elevation from 15 >> > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't expect a >> > big difference in performance, but that is the standard). The rain >> > gauge >> > should be similarly sited AGL to minimize potential under-catch caused >> > by turbulent wind flow across the top of the rain gauge from stronger >> > winds at the higher siting elevation. >> > >> > Dave >> > CW0351 >> > >> > >> > >> > Paul Grace wrote: >> > >> > > The CWOP info for my station (CW0622) is showing an "error" of -4.1 >> > > ?F, the worst average error of -16.4. >> > > >> > > In looking at the neighboring stations, they are all in very >> > > different >> > > areas- I am on a 640m ridge 10 miles from Pt. Arena, California, >> > > while >> > > my "neighbors" are 20 - 60 miles east, in warmer valleys a few meters >> > > above sea level. I would expect lower temperatures at my station. I >> > > have a Davis fan-aspirated temp. sensor, that seems correct to within >> > > a fraction of a degree when compared to a thermocouple. It is >> > > situated >> > > in the middle of a small field, about 15ft off the ground, a couple >> > > hundred feet from the nearest trees. >> > > >> > > How are the "correct" temperature readings >> > > interpolated/extrapolated/measured? Is my station being compared to >> > > these "neighbors"? >> > > >> > >------------------------------------------------------------------------ >> > > >> > >_______________________________________________ >> > >wxqc mailing list >> > >wxqc@lists.gladstonefamily.net >> > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > > >> > >The contents of this message are the responsibility of the author. >> > > >> > >> > _______________________________________________ >> > wxqc mailing list >> > wxqc@lists.gladstonefamily.net >> > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> > >> > The contents of this message are the responsibility of the author. >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. >> >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. From dshelms at comcast.net Tue Nov 22 00:16:34 2005 From: dshelms at comcast.net (Dave Helms) Date: Tue Nov 22 00:16:38 2005 Subject: [wxqc] Temperature error In-Reply-To: <004201c5ef20$437ee9f0$6401a8c0@intel3200> References: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> <004201c5ef20$437ee9f0$6401a8c0@intel3200> Message-ID: <4382A9B2.3040801@comcast.net> Hi Jack, My latest posts concerning MADIS QCMS had nothing to do with numerical models nor climate trends. In response to questions posted to the list, I addressed aspects of the QCMS statistical analysis of observational data which does a nice job for most stations most of the time. Now lets please limit discussion to CWOP and weather sensors and observing techniques... and try to maintain mutual respect in our posts. If this is not possible, maybe you should consider unsubscribing from this list. Regards, Dave Jack Jones wrote: > Hi Dave, > > You weather guys are a "trip". If I had been as wrong as often as > weather forecasters in my previous job, I would have been unemployed > for the last 30 years. However, weather forecasters hve the audacity > to present a model to forecast temperatures & other weather data, then > "cry foul" when someone presents them with data that is different from > the forecast. How many times have I carried an umbrella on sunny days > just because a weather forecaster said it will rain. > > Jack Jones > ----- Original Message ----- From: > To: "Discussion of weather data quality issues" > ; "'Discussion of weather data quality > issues'" > Sent: Monday, November 21, 2005 4:15 PM > Subject: RE: [wxqc] Temperature error > > >> Hi Paul, >> >> Your data are not really flagged as bad, just something that needs >> greater attention to see what the real deal is. Because there are >> very few observations from your area is a great reason for you to >> continue to contributing your observations. Ultimately, the local >> weather guys down at Monterrey and up in Eureka know from looking at >> the station data what is good and not so good. >> >> Again, MADIS QCMS is a first guess and not a final statement of quality. >> >> Cheers, >> >> Dave >> >> >>> Yes, we do look out on a river valley often full of fog, we are almost >>> always above it. >>> So you're a "slow" weather weenie? Scary. >>> >>> For a big picture question; is the fact that QCMS tags my >>> temps/dewpoint as >>> bad make any difference with the use of my data? Is my data used >>> for any >>> actual purpose? There are very few stations near mine, so I thought it >>> might be a different data point, but if it's tossed out, then should I >>> bother uploading it? >>> >>> -----Original Message----- >>> From: wxqc-bounces@lists.gladstonefamily.net >>> [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of >>> dshelms@comcast.net >>> Sent: Monday, November 21, 2005 11:20 >>> To: Discussion of weather data quality issues; Discussion of weather >>> data >>> quality issues >>> Subject: Re: [wxqc] Temperature error >>> >>> FYI... Just to do a little Weather 101 (apologies in advance for the >>> techno >>> stuff) >>> >>> I probably simplified the situation along the northern California >>> coast in >>> my previous discussion. I was "slicing" the temperatures in your >>> area this >>> morning using MADIS mesonet page and noticed stations above 1,500 ft >>> (500 >>> meters) AMSL were in the 60's F (15-ish degrees C) while the >>> stations below >>> 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). >>> >>> A west coast weather guy would recognize that as a classic subsidence >>> inversion (temperatures warm with height), but I am an east coast >>> weather >>> weanie (and a little slow, I'm afraid). Checking out today's >>> Oakland, CA, >>> sounding from the NWS weather balloon (aka radiosonde): >>> http://weather.unisys.com/upper_air/skew/skew_KOAK.html >>> FYI... the left-hand white line is the dew point trace, where the >>> temperature and dew point converge, its wet; where they diverge, its >>> dry. >>> I'm assuming you are often looking down at lots of fog and low >>> clouds from >>> your ranch as a result of the subsidense inversion you live in/near. >>> >>> The diagnal solid blue lines are the temperature isotherms (e.g. >>> "skewed" >>> T), you see the right hand white line (atmospheric temperature) starts >>> around 1000 mb (177 m or meters) at about 8 C degrees and increases >>> with >>> height until it tops out at almost 20 C at about 800 meters (about >>> 2,400 >>> ft). So... more than decoupling, your station (and others between >>> 2-3,000 >>> ft near you) trace the warm air near the maximum temperature of the >>> subsidence inversion! >>> >>> Again, MADIS QCMS assumes the atmosphere cools with height which is >>> not true >>> for your location in many cases. Thus, the inadvertent "flag" of your >>> temperature. >>> >>> >>> Mystery solved! >>> >>> Dave >>> >>> >>> > Hi Paul, >>> > >>> > MADIS QCMS uses a potential temperature comparison which corrects for >>> > elevation differences (dry adiabatic lapse rate) to assess station >>> > performance. Mesowest does a similar temperature comparison > >>> methodology, >>> > and also appears to show your temperatures running a little warm, >>> after >>> > correcting for elevation: >>> > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 >>> > >>> > We have seen that valley stations run a bit "cool", and mountaintop >>> > stations, such as yours, run a bit "warm" in these statistical > >>> analyzes. >>> > We believe these warm/cool bias anomalies appear in the statistics >>> > because the algorithm does not account the amount of atmospheric >>> mixing >>> > occurring at individual sites. Valley sites tend to "de-couple" at >>> > night >>> > with little atmospheric mixing and added cooling from radiational >>> > cooling and drainage winds. Mountaintop stations are well mixed >>> > throughout the night and don't typically have evening radiational >>> > cooling to the degree seen in valley stations. >>> > >>> > Given that you have an active aspirated temperature sensor sited >>> in an >>> > open area, I think that your temperature observations are > >>> representative >>> > for your location. A little nit... I would lower the elevation >>> from 15 >>> > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't >>> expect a >>> > big difference in performance, but that is the standard). The rain >>> > gauge >>> > should be similarly sited AGL to minimize potential under-catch >>> caused >>> > by turbulent wind flow across the top of the rain gauge from stronger >>> > winds at the higher siting elevation. >>> > >>> > Dave >>> > CW0351 >>> > >>> > >>> > >>> > Paul Grace wrote: >>> > >>> > > The CWOP info for my station (CW0622) is showing an "error" of -4.1 >>> > > ?F, the worst average error of -16.4. >>> > > >>> > > In looking at the neighboring stations, they are all in very > > >>> different >>> > > areas- I am on a 640m ridge 10 miles from Pt. Arena, California, >>> > > while >>> > > my "neighbors" are 20 - 60 miles east, in warmer valleys a few >>> meters >>> > > above sea level. I would expect lower temperatures at my station. I >>> > > have a Davis fan-aspirated temp. sensor, that seems correct to >>> within >>> > > a fraction of a degree when compared to a thermocouple. It is > >>> > situated >>> > > in the middle of a small field, about 15ft off the ground, a couple >>> > > hundred feet from the nearest trees. >>> > > >>> > > How are the "correct" temperature readings >>> > > interpolated/extrapolated/measured? Is my station being compared to >>> > > these "neighbors"? >>> > > >>> > >>> >------------------------------------------------------------------------ >>> >>> > > >>> > >_______________________________________________ >>> > >wxqc mailing list >>> > >wxqc@lists.gladstonefamily.net >>> > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> > > >>> > >The contents of this message are the responsibility of the author. >>> > > >>> > >>> > _______________________________________________ >>> > wxqc mailing list >>> > wxqc@lists.gladstonefamily.net >>> > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> > >>> > The contents of this message are the responsibility of the author. >>> _______________________________________________ >>> wxqc mailing list >>> wxqc@lists.gladstonefamily.net >>> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> >>> The contents of this message are the responsibility of the author. >>> >>> _______________________________________________ >>> wxqc mailing list >>> wxqc@lists.gladstonefamily.net >>> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> >>> The contents of this message are the responsibility of the author. >> >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. > > > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. > > From sam at wa4phy.net Tue Nov 22 00:21:31 2005 From: sam at wa4phy.net (Sam Drinkard) Date: Tue Nov 22 00:21:35 2005 Subject: [wxqc] Temperature error In-Reply-To: <004201c5ef20$437ee9f0$6401a8c0@intel3200> References: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> <004201c5ef20$437ee9f0$6401a8c0@intel3200> Message-ID: <4382AADB.6080608@wa4phy.net> Jack Jones wrote: > Hi Dave, > > You weather guys are a "trip". If I had been as wrong as often as > weather forecasters in my previous job, I would have been unemployed > for the last 30 years. However, weather forecasters hve the audacity > to present a model to forecast temperatures & other weather data, then > "cry foul" when someone presents them with data that is different from > the forecast. How many times have I carried an umbrella on sunny days > just because a weather forecaster said it will rain. > > Jack Jones > Jack, The numerical models are just another "tool" that helps the meteorologist decide what is *going* to happen vs. what IS happening. There is some skill in the models, but the real job of generating a forecast is left up to the skill of the person using all the available data. As Dave pointed out, your numbers are not "bad" -- just different because of your location. I'm in such a location myself. If I compare my data to AGS, I'd always be off, because AGS is down in the hole along side the Savannah river, at an elevation of 147', where I am at 400+'. Those differences in elevation are smoothed out by converting the temperature you send in to a potential temperature, just as Dave explained (and I think he did a great job of presenting the skew-t plot too). It sound to me like perhaps you have gotten some wrong information somewhere about what takes place. Check out the MADIS QC information pages, and if you really want to see what the models are all about with their weaknesses and strong points, try this URL to get a better handle on things. As for the elevation change, as you know, altitude does have an effect on temperature. 10' or 15' might not sound like much, but believe me, it can make a big difference on those nights when the atmosphere decouples very close to the ground, but you still have some mixing at a big higher elevation. http://www.hpc.ncep.noaa.gov/html/model2.shtml Just about any kind of information you'd like to know about the numerical models. Very good reading, and *IF* you have some type of information that you think is a bias towards one model or the other, HPC welcomes input from its visitors and users. cheers! Sam -- Snowman From sniadoch at frontiernet.net Tue Nov 22 00:25:01 2005 From: sniadoch at frontiernet.net (Hank Sniadoch) Date: Tue Nov 22 00:26:18 2005 Subject: [wxqc] Temperature error In-Reply-To: <4382A9B2.3040801@comcast.net> References: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> <004201c5ef20$437ee9f0$6401a8c0@intel3200> <4382A9B2.3040801@comcast.net> Message-ID: <4382ABAD.9020800@frontiernet.net> Good going !! Dave Helms wrote: > Hi Jack, > > My latest posts concerning MADIS QCMS had nothing to do with numerical > models nor climate trends. In response to questions posted to the > list, I addressed aspects of the QCMS statistical analysis of > observational data which does a nice job for most stations most of the > time. > Now lets please limit discussion to CWOP and weather sensors and > observing techniques... and try to maintain mutual respect in our > posts. If this is not possible, maybe you should consider > unsubscribing from this list. > > > Regards, > > Dave > > Jack Jones wrote: > >> Hi Dave, >> >> You weather guys are a "trip". If I had been as wrong as often as >> weather forecasters in my previous job, I would have been unemployed >> for the last 30 years. However, weather forecasters hve the audacity >> to present a model to forecast temperatures & other weather data, >> then "cry foul" when someone presents them with data that is >> different from the forecast. How many times have I carried an >> umbrella on sunny days just because a weather forecaster said it >> will rain. >> >> Jack Jones >> ----- Original Message ----- From: >> To: "Discussion of weather data quality issues" >> ; "'Discussion of weather data >> quality issues'" >> Sent: Monday, November 21, 2005 4:15 PM >> Subject: RE: [wxqc] Temperature error >> >> >>> Hi Paul, >>> >>> Your data are not really flagged as bad, just something that needs >>> greater attention to see what the real deal is. Because there are >>> very few observations from your area is a great reason for you to >>> continue to contributing your observations. Ultimately, the local >>> weather guys down at Monterrey and up in Eureka know from looking at >>> the station data what is good and not so good. >>> >>> Again, MADIS QCMS is a first guess and not a final statement of >>> quality. >>> >>> Cheers, >>> >>> Dave >>> >>> >>>> Yes, we do look out on a river valley often full of fog, we are almost >>>> always above it. >>>> So you're a "slow" weather weenie? Scary. >>>> >>>> For a big picture question; is the fact that QCMS tags my >>>> temps/dewpoint as >>>> bad make any difference with the use of my data? Is my data used >>>> for any >>>> actual purpose? There are very few stations near mine, so I >>>> thought it >>>> might be a different data point, but if it's tossed out, then should I >>>> bother uploading it? >>>> >>>> -----Original Message----- >>>> From: wxqc-bounces@lists.gladstonefamily.net >>>> [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of >>>> dshelms@comcast.net >>>> Sent: Monday, November 21, 2005 11:20 >>>> To: Discussion of weather data quality issues; Discussion of >>>> weather data >>>> quality issues >>>> Subject: Re: [wxqc] Temperature error >>>> >>>> FYI... Just to do a little Weather 101 (apologies in advance for >>>> the techno >>>> stuff) >>>> >>>> I probably simplified the situation along the northern California >>>> coast in >>>> my previous discussion. I was "slicing" the temperatures in your >>>> area this >>>> morning using MADIS mesonet page and noticed stations above 1,500 >>>> ft (500 >>>> meters) AMSL were in the 60's F (15-ish degrees C) while the >>>> stations below >>>> 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). >>>> >>>> A west coast weather guy would recognize that as a classic subsidence >>>> inversion (temperatures warm with height), but I am an east coast >>>> weather >>>> weanie (and a little slow, I'm afraid). Checking out today's >>>> Oakland, CA, >>>> sounding from the NWS weather balloon (aka radiosonde): >>>> http://weather.unisys.com/upper_air/skew/skew_KOAK.html >>>> FYI... the left-hand white line is the dew point trace, where the >>>> temperature and dew point converge, its wet; where they diverge, >>>> its dry. >>>> I'm assuming you are often looking down at lots of fog and low >>>> clouds from >>>> your ranch as a result of the subsidense inversion you live in/near. >>>> >>>> The diagnal solid blue lines are the temperature isotherms (e.g. >>>> "skewed" >>>> T), you see the right hand white line (atmospheric temperature) starts >>>> around 1000 mb (177 m or meters) at about 8 C degrees and increases >>>> with >>>> height until it tops out at almost 20 C at about 800 meters (about >>>> 2,400 >>>> ft). So... more than decoupling, your station (and others between >>>> 2-3,000 >>>> ft near you) trace the warm air near the maximum temperature of the >>>> subsidence inversion! >>>> >>>> Again, MADIS QCMS assumes the atmosphere cools with height which is >>>> not true >>>> for your location in many cases. Thus, the inadvertent "flag" of your >>>> temperature. >>>> >>>> >>>> Mystery solved! >>>> >>>> Dave >>>> >>>> >>>> > Hi Paul, >>>> > >>>> > MADIS QCMS uses a potential temperature comparison which corrects >>>> for >>>> > elevation differences (dry adiabatic lapse rate) to assess station >>>> > performance. Mesowest does a similar temperature comparison > >>>> methodology, >>>> > and also appears to show your temperatures running a little warm, >>>> after >>>> > correcting for elevation: >>>> > >>>> http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 >>>> > >>>> > We have seen that valley stations run a bit "cool", and mountaintop >>>> > stations, such as yours, run a bit "warm" in these statistical > >>>> analyzes. >>>> > We believe these warm/cool bias anomalies appear in the statistics >>>> > because the algorithm does not account the amount of atmospheric >>>> mixing >>>> > occurring at individual sites. Valley sites tend to "de-couple" >>>> at > night >>>> > with little atmospheric mixing and added cooling from radiational >>>> > cooling and drainage winds. Mountaintop stations are well mixed >>>> > throughout the night and don't typically have evening radiational >>>> > cooling to the degree seen in valley stations. >>>> > >>>> > Given that you have an active aspirated temperature sensor sited >>>> in an >>>> > open area, I think that your temperature observations are > >>>> representative >>>> > for your location. A little nit... I would lower the elevation >>>> from 15 >>>> > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't >>>> expect a >>>> > big difference in performance, but that is the standard). The >>>> rain > gauge >>>> > should be similarly sited AGL to minimize potential under-catch >>>> caused >>>> > by turbulent wind flow across the top of the rain gauge from >>>> stronger >>>> > winds at the higher siting elevation. >>>> > >>>> > Dave >>>> > CW0351 >>>> > >>>> > >>>> > >>>> > Paul Grace wrote: >>>> > >>>> > > The CWOP info for my station (CW0622) is showing an "error" of >>>> -4.1 >>>> > > ?F, the worst average error of -16.4. >>>> > > >>>> > > In looking at the neighboring stations, they are all in very > >>>> > different >>>> > > areas- I am on a 640m ridge 10 miles from Pt. Arena, >>>> California, > > while >>>> > > my "neighbors" are 20 - 60 miles east, in warmer valleys a few >>>> meters >>>> > > above sea level. I would expect lower temperatures at my >>>> station. I >>>> > > have a Davis fan-aspirated temp. sensor, that seems correct to >>>> within >>>> > > a fraction of a degree when compared to a thermocouple. It is > >>>> > situated >>>> > > in the middle of a small field, about 15ft off the ground, a >>>> couple >>>> > > hundred feet from the nearest trees. >>>> > > >>>> > > How are the "correct" temperature readings >>>> > > interpolated/extrapolated/measured? Is my station being >>>> compared to >>>> > > these "neighbors"? >>>> > > >>>> > >>>> >------------------------------------------------------------------------ >>>> >>>> > > >>>> > >_______________________________________________ >>>> > >wxqc mailing list >>>> > >wxqc@lists.gladstonefamily.net >>>> > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>>> > > >>>> > >The contents of this message are the responsibility of the author. >>>> > > >>>> > >>>> > _______________________________________________ >>>> > wxqc mailing list >>>> > wxqc@lists.gladstonefamily.net >>>> > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>>> > >>>> > The contents of this message are the responsibility of the author. >>>> _______________________________________________ >>>> wxqc mailing list >>>> wxqc@lists.gladstonefamily.net >>>> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>>> >>>> The contents of this message are the responsibility of the author. >>>> >>>> _______________________________________________ >>>> wxqc mailing list >>>> wxqc@lists.gladstonefamily.net >>>> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>>> >>>> The contents of this message are the responsibility of the author. >>> >>> >>> _______________________________________________ >>> wxqc mailing list >>> wxqc@lists.gladstonefamily.net >>> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> >>> The contents of this message are the responsibility of the author. >> >> >> >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. >> >> > > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. > > From tim.mcmanus at mac.com Tue Nov 22 00:26:54 2005 From: tim.mcmanus at mac.com (tim.mcmanus@mac.com) Date: Tue Nov 22 00:27:04 2005 Subject: [wxqc] Dew Point In-Reply-To: <200511220448.jAM4m6Er014834@mac.com> References: <200511220448.jAM4m6Er014834@mac.com> Message-ID: I have an issue with an inaccurate dew point. I originally had my fan-aspirated Vantage Pro on the roof of my house during the summer months. I then moved it off of the roof and onto a post in the back yard approximately 5ft. over grass. The post is a wooden post. After the move, the dew point read 3.2 ?F higher than the statistical model. i don't understand why. I assumed that the move would bring the station readings more in line with the model, but it has deviated but only in this one area. It might be due to the seasonal change and how it interacts with the terrain because I moved the station October 1. The sun also hits the station later in the morning, and that might be screwing with things. I also have a leaf wetness sensor to correlate high periods of humidity. In some instances my humidity readings will be 10%-20% higher than C3708 or AP752, both within 3 miles of my location. The station sits along the bottom of a narrow valley that's about a mile across. It's not really a valley per se, but it's the best description I have. I haven't had a chance to borrow a sling to compare humidity readings, but I am tending to believe that they are correct. Suggestions? Thanks! -Tim McManus tim.mcmanus at mac.com From sniadoch at frontiernet.net Tue Nov 22 00:25:52 2005 From: sniadoch at frontiernet.net (Hank Sniadoch) Date: Tue Nov 22 00:27:09 2005 Subject: [wxqc] Temperature error In-Reply-To: <004201c5ef20$437ee9f0$6401a8c0@intel3200> References: <112120052215.6585.438246E4000D7E17000019B922007348309C03040A089C0B@comcast.net> <004201c5ef20$437ee9f0$6401a8c0@intel3200> Message-ID: <4382ABE0.7080406@frontiernet.net> Who's a trip ?? Them ?? I think not .... maybe you should look in the mirror. Jack Jones wrote: > Hi Dave, > > You weather guys are a "trip". If I had been as wrong as often as > weather forecasters in my previous job, I would have been unemployed > for the last 30 years. However, weather forecasters hve the audacity > to present a model to forecast temperatures & other weather data, then > "cry foul" when someone presents them with data that is different from > the forecast. How many times have I carried an umbrella on sunny days > just because a weather forecaster said it will rain. > > Jack Jones > ----- Original Message ----- From: > To: "Discussion of weather data quality issues" > ; "'Discussion of weather data quality > issues'" > Sent: Monday, November 21, 2005 4:15 PM > Subject: RE: [wxqc] Temperature error > > >> Hi Paul, >> >> Your data are not really flagged as bad, just something that needs >> greater attention to see what the real deal is. Because there are >> very few observations from your area is a great reason for you to >> continue to contributing your observations. Ultimately, the local >> weather guys down at Monterrey and up in Eureka know from looking at >> the station data what is good and not so good. >> >> Again, MADIS QCMS is a first guess and not a final statement of quality. >> >> Cheers, >> >> Dave >> >> >>> Yes, we do look out on a river valley often full of fog, we are almost >>> always above it. >>> So you're a "slow" weather weenie? Scary. >>> >>> For a big picture question; is the fact that QCMS tags my >>> temps/dewpoint as >>> bad make any difference with the use of my data? Is my data used >>> for any >>> actual purpose? There are very few stations near mine, so I thought it >>> might be a different data point, but if it's tossed out, then should I >>> bother uploading it? >>> >>> -----Original Message----- >>> From: wxqc-bounces@lists.gladstonefamily.net >>> [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of >>> dshelms@comcast.net >>> Sent: Monday, November 21, 2005 11:20 >>> To: Discussion of weather data quality issues; Discussion of weather >>> data >>> quality issues >>> Subject: Re: [wxqc] Temperature error >>> >>> FYI... Just to do a little Weather 101 (apologies in advance for the >>> techno >>> stuff) >>> >>> I probably simplified the situation along the northern California >>> coast in >>> my previous discussion. I was "slicing" the temperatures in your >>> area this >>> morning using MADIS mesonet page and noticed stations above 1,500 ft >>> (500 >>> meters) AMSL were in the 60's F (15-ish degrees C) while the >>> stations below >>> 1,500 ft AMSL were in the 40s-50s (5-13 degrees C). >>> >>> A west coast weather guy would recognize that as a classic subsidence >>> inversion (temperatures warm with height), but I am an east coast >>> weather >>> weanie (and a little slow, I'm afraid). Checking out today's >>> Oakland, CA, >>> sounding from the NWS weather balloon (aka radiosonde): >>> http://weather.unisys.com/upper_air/skew/skew_KOAK.html >>> FYI... the left-hand white line is the dew point trace, where the >>> temperature and dew point converge, its wet; where they diverge, its >>> dry. >>> I'm assuming you are often looking down at lots of fog and low >>> clouds from >>> your ranch as a result of the subsidense inversion you live in/near. >>> >>> The diagnal solid blue lines are the temperature isotherms (e.g. >>> "skewed" >>> T), you see the right hand white line (atmospheric temperature) starts >>> around 1000 mb (177 m or meters) at about 8 C degrees and increases >>> with >>> height until it tops out at almost 20 C at about 800 meters (about >>> 2,400 >>> ft). So... more than decoupling, your station (and others between >>> 2-3,000 >>> ft near you) trace the warm air near the maximum temperature of the >>> subsidence inversion! >>> >>> Again, MADIS QCMS assumes the atmosphere cools with height which is >>> not true >>> for your location in many cases. Thus, the inadvertent "flag" of your >>> temperature. >>> >>> >>> Mystery solved! >>> >>> Dave >>> >>> >>> > Hi Paul, >>> > >>> > MADIS QCMS uses a potential temperature comparison which corrects for >>> > elevation differences (dry adiabatic lapse rate) to assess station >>> > performance. Mesowest does a similar temperature comparison > >>> methodology, >>> > and also appears to show your temperatures running a little warm, >>> after >>> > correcting for elevation: >>> > http://www.met.utah.edu/cgi-bin/droman/regress_database.cgi?stn=C0622 >>> > >>> > We have seen that valley stations run a bit "cool", and mountaintop >>> > stations, such as yours, run a bit "warm" in these statistical > >>> analyzes. >>> > We believe these warm/cool bias anomalies appear in the statistics >>> > because the algorithm does not account the amount of atmospheric >>> mixing >>> > occurring at individual sites. Valley sites tend to "de-couple" at >>> > night >>> > with little atmospheric mixing and added cooling from radiational >>> > cooling and drainage winds. Mountaintop stations are well mixed >>> > throughout the night and don't typically have evening radiational >>> > cooling to the degree seen in valley stations. >>> > >>> > Given that you have an active aspirated temperature sensor sited >>> in an >>> > open area, I think that your temperature observations are > >>> representative >>> > for your location. A little nit... I would lower the elevation >>> from 15 >>> > ft AGL to 5 ft AGL for the temperature/RH sensors (I wouldn't >>> expect a >>> > big difference in performance, but that is the standard). The rain >>> > gauge >>> > should be similarly sited AGL to minimize potential under-catch >>> caused >>> > by turbulent wind flow across the top of the rain gauge from stronger >>> > winds at the higher siting elevation. >>> > >>> > Dave >>> > CW0351 >>> > >>> > >>> > >>> > Paul Grace wrote: >>> > >>> > > The CWOP info for my station (CW0622) is showing an "error" of -4.1 >>> > > ?F, the worst average error of -16.4. >>> > > >>> > > In looking at the neighboring stations, they are all in very > > >>> different >>> > > areas- I am on a 640m ridge 10 miles from Pt. Arena, California, >>> > > while >>> > > my "neighbors" are 20 - 60 miles east, in warmer valleys a few >>> meters >>> > > above sea level. I would expect lower temperatures at my station. I >>> > > have a Davis fan-aspirated temp. sensor, that seems correct to >>> within >>> > > a fraction of a degree when compared to a thermocouple. It is > >>> > situated >>> > > in the middle of a small field, about 15ft off the ground, a couple >>> > > hundred feet from the nearest trees. >>> > > >>> > > How are the "correct" temperature readings >>> > > interpolated/extrapolated/measured? Is my station being compared to >>> > > these "neighbors"? >>> > > >>> > >>> >------------------------------------------------------------------------ >>> >>> > > >>> > >_______________________________________________ >>> > >wxqc mailing list >>> > >wxqc@lists.gladstonefamily.net >>> > >http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> > > >>> > >The contents of this message are the responsibility of the author. >>> > > >>> > >>> > _______________________________________________ >>> > wxqc mailing list >>> > wxqc@lists.gladstonefamily.net >>> > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> > >>> > The contents of this message are the responsibility of the author. >>> _______________________________________________ >>> wxqc mailing list >>> wxqc@lists.gladstonefamily.net >>> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> >>> The contents of this message are the responsibility of the author. >>> >>> _______________________________________________ >>> wxqc mailing list >>> wxqc@lists.gladstonefamily.net >>> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >>> >>> The contents of this message are the responsibility of the author. >> >> _______________________________________________ >> wxqc mailing list >> wxqc@lists.gladstonefamily.net >> http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc >> >> The contents of this message are the responsibility of the author. > > > _______________________________________________ > wxqc mailing list > wxqc@lists.gladstonefamily.net > http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc > > The contents of this message are the responsibility of the author. > > From sam at wa4phy.net Tue Nov 22 00:33:02 2005 From: sam at wa4phy.net (Sam Drinkard) Date: Tue Nov 22 00:33:04 2005 Subject: [wxqc] Rain Gauge undercatch Message-ID: <4382AD8E.8040002@wa4phy.net> Hi List, Some time back, I vaguely recall reading something, or following a link someone provided that gave a breakdown or the mechanics of undercatch with various shapes of rain gauges, plus the tipping bucket type. If any of you happen to remember the URL or discussion, I'd appreciate a hint on where it is.. Saw quite well this afternoon what and how bad undercatch with the Davis tipping bucket gauge can be. We had, according to the WM-II, a precip rate of about .74"/hr. I've not checked the SRG yet but I know my backup 4" gauge was showing way more rain than the Davis recorded, and I've already tweaked the adjustments as far as they will go towards the plus side. -- Snowman From tim.mcmanus at mac.com Tue Nov 22 00:37:57 2005 From: tim.mcmanus at mac.com (tim.mcmanus@mac.com) Date: Tue Nov 22 00:38:05 2005 Subject: [wxqc] Forgot to identify my station... In-Reply-To: <200511220534.jAM5Y6sp006078@mac.com> References: <200511220534.jAM5Y6sp006078@mac.com> Message-ID: <0CCA19D1-3267-4B09-B36E-343102E7D075@mac.com> Whoops, I forgot to identify my station. It's CW4023. ;-) -Tim McManus tim.mcmanus at mac.com From rich.taft at att.net Tue Nov 22 00:57:05 2005 From: rich.taft at att.net (rich.taft@att.net) Date: Tue Nov 22 00:57:09 2005 Subject: [wxqc] Pressure Error Message-ID: <112220050557.17695.4382B33100087F0F0000451F21603762239B000E9BD2080C079D@att.net> I own a Davis Vantage Pro 2 Plus for almost 9 months now. I have overall been very happy with the accuracy, but I have notice a small glitch over the last few months. I keep getting QC errors with my pressure reading. Its not a consistent error, sometimes it only 4 or five errors out of 96 samples, Sometimes all 96 fail, other days no errors at all. Sometimes the station reports pressure too high, sometimes too low. I have made a couple of attempts to adjust the console by small increments. This only fixes it for a couple of days and the error creeps back. I called Davis and told them my findings and they very nicely replaced my console. Again it only seemed to be fixed for a few days before the error creeps back. The stations around me are relatively close, so I would expect my readings to be close. It seems to have something to do with temperature, or more accurately the direction it is heading and how fast, but since the sensor is indoors I know its not being affected. I did read someplace that Davis adjusts the raw pressure based on temperature & altitude, not just on altitude as a pilot would. I'll be the first one to admit that I'm no expert, and have learned most of what I know about weather data collection as I go along. I'm looking for a possible answer from someone that knows better then I (which is just about anyone). I know I have my location and altitude properly set in the console, and have checked it several times. Maybe someone can point me in the right direction to "clean up" my data. -- Richard Taft www.taftphoto.com From gary.oldham at adelphia.net Tue Nov 22 08:16:10 2005 From: gary.oldham at adelphia.net (Gary Oldham) Date: Tue Nov 22 08:16:20 2005 Subject: [wxqc] Pressure Error Message-ID: <16671103.1132665370736.JavaMail.root@web6.mail.adelphia.net> Rich - As far as I know - and I'm moderately knowledgeable on the VP - Davis does not use temperature compensation, only altitude. There's likely to be a certain amount of "drift," but if you're at a similar altitude to the nearby stations to which you're comparing, I'd expect you to be relatively close. If your console is placed somewhere subject to a lot of temperature fluctuations (like, for instance, near a heater or AC vent), that could certainly cause anomalies in your readings. A couple of other things to consider. If you're using airport METAR data in your comparisons, see how often they update. Some airports only update once per hour. The VantagePro, unfortunately, only updates every fifteen minutes, a design flaw I'd hoped would have been changed in the VP2 but wasn't. I have both a VP1 and VP2, and their consoles are very near one another, and the barometer readings on the two units diverge quite a bit sometimes. There are good and comprehensive instructions on how to calibrate your barometer in the CWOP Instrument Siting Guide, which can be downloaded at http://mywebpages.comcast.net/dshelms/CWOP_Guide.pdf. There are many other good tips in there on placement of your instruments for optimal accuracy (including, if I remember right, some good disucssion on rain gauge catchment issues as asked in someone else's recent post. Jack - you might want to consider setting up a blog as a venue for your rants. There are many free ones available, and with a blog, only those who actually *want* to read your rants can do so, and those that don't want to see them clogging up this weather data quality discussion list won't have to. I, for one, appreciate the insights offered by the professionals who generously donate their time to this program. I view all input, expert or otherwise, with a critical eye, and accept it or not based on many factors, but don't denigrate those who offer constructive suggestions, even if I choose not to heed it. And please don't compare Joe Bastardi to the professionals at the NWS. Bastardi is a showman, a self-aggrandizing buffoon who is quite often wrong but will end up spinning his guesscasts after the fact to make it look like he called it right on the nose. The NHC did a superb job during this unprecedented hurricane season, and without hype or nonsense. The best forecasters augment the models with their own knowledge of local terrain, microclimates, and other geophysical elements, but they're still going to be wrong sometimes. Those that strictly follow the models without human augmentation are probably going to be wrong more often than not. If you think of trying to predict 72 or more hours out (or 24 or less for that matter) what's going to happen on a huge, grossly irregular sphere, covered in a fluid of constantly changing density, with literally millions of ever-evolving factors, those prognostications are going to be wrong sometimes. Probably a lot. The quality of forecasts coming out of NWS have steadily improved over the years, and today, a 72 or even 96 hour forecast is going to be as accurate as a 24 hour forecast only a couple of decades ago. gary.oldham(at)adelphia.net ---- rich.taft@att.net wrote: ============= I own a Davis Vantage Pro 2 Plus for almost 9 months now. I have overall been very happy with the accuracy, but I have notice a small glitch over the last few months. I keep getting QC errors with my pressure reading. Its not a consistent error, sometimes it only 4 or five errors out of 96 samples, Sometimes all 96 fail, other days no errors at all. Sometimes the station reports pressure too high, sometimes too low. I have made a couple of attempts to adjust the console by small increments. This only fixes it for a couple of days and the error creeps back. I called Davis and told them my findings and they very nicely replaced my console. Again it only seemed to be fixed for a few days before the error creeps back. The stations around me are relatively close, so I would expect my readings to be close. It seems to have something to do with temperature, or more accurately the direction it is heading and how fast, but since the sensor is indoors I know its not being affected. I did read someplace that Davis adjusts the raw pressure based on temperature & altitude, not just on altitude as a pilot would. I'll be the first one to admit that I'm no expert, and have learned most of what I know about weather data collection as I go along. I'm looking for a possible answer from someone that knows better then I (which is just about anyone). I know I have my location and altitude properly set in the console, and have checked it several times. Maybe someone can point me in the right direction to "clean up" my data. -- Richard Taft www.taftphoto.com _______________________________________________ wxqc mailing list wxqc@lists.gladstonefamily.net http://pond1.gladstonefamily.net:8080/mailman/listinfo/wxqc The contents of this message are the responsibility of the author. From david at coursey.com Tue Nov 22 01:04:53 2005 From: david at coursey.com (David Coursey) Date: Tue Nov 22 09:00:20 2005 Subject: [wxqc] Temperature error Message-ID: <2AA5D5B18E1F4849A2B998B43BAA2BA70FC70A@fjuckby.coursey.local> Thanks for your assistance. Just wanted to make sure the station was working properly. Ziff Davis Internet David E. Coursey Editor-at-Large david@coursey.com david_coursey@ziffdavis.com 1528 Tamarisk Lane Tracy, CA 95377-8273 tel: (209) 835-6893 mobile: (209) 740-7515 IM: david_coursey (YIM) www.coursey.com ? Add me to your address book... Want a signature like this?? -----Original Message----- From: wxqc-bounces@lists.gladstonefamily.net [mailto:wxqc-bounces@lists.gladstonefamily.net] On Behalf Of Dave Helm