skierinvermont Posted February 7, 2011 Share Posted February 7, 2011 Ok, my guess was correct. RSS is intended to report on raw channel 2 data. RSS was initially developed in 2001 in this paper by Mears et al. It was intended as an independent complementary analysis to that done by Christy and Spencer at UAH. http://www.ncdc.noaa.gov/temp-and-precip/msu/rss-msu.pdf In that paper they state that they are reporting on raw channel 2 data, which contains a stratospheric contribution. They don't correct for it, because at the time they did not think the methodologies for doing so were sound. Link to comment Share on other sites More sharing options...
skierinvermont Posted February 7, 2011 Share Posted February 7, 2011 Hmm...guess I'm more confused. I thought the published UAH and RSS anomalies were supposed to represent the LT (lower troposphere), not the mid-troposphere. I can see how the stratosphere would effect the middle tropopsheric numbers a little, but not so much the lower troposphere. In addition, I find the phrasing in that last paragraph a bit odd: the adjustment is an "attempt to remove the cooling influence". Makes it sound like Fu set out to "correct" things warmer, rather than just apply a better analysis. I believe it is because they are simply broadcasting at a frequency downwards from space. Channel 2 is at 53.74 Ghz. Because the beam must pass through the stratosphere, naturally that would affect the results. Everybody seems to agree there is a stratospheric contribution, even Spencer and Christy, the only difference is on how to correct for it. It seems like the many papers published in the last 5 years have come to an agreement, with Spencer and Christy in a small minority that disagrees. Link to comment Share on other sites More sharing options...
tacoman25 Posted February 7, 2011 Share Posted February 7, 2011 Ok, my guess was correct. RSS is intended to report on raw channel 2 data. RSS was initially developed in 2001 in this paper by Mears et al. It was intended as an independent complementary analysis to that done by Christy and Spencer at UAH. http://www.ncdc.noaa...msu/rss-msu.pdf In that paper they state that they are reporting on raw channel 2 data, which contains a stratospheric contribution. They don't correct for it, because at the time they did not think the methodologies for doing so were sound. Correct me if I'm wrong, but that is different than UAH, right? I believe UAH is much closer to the Ch. 5 data....which I assume is different than Ch. 2. Also, if you see my previous post, I'm confused at to why the mid-troposphere is being referenced, when RSS and UAH anomalies were both supposed to be LT temperatures... Link to comment Share on other sites More sharing options...
tacoman25 Posted February 7, 2011 Share Posted February 7, 2011 I believe it is because they are simply broadcasting at a frequency downwards from space. Channel 2 is at 53.74 Ghz. Because the beam must pass through the stratosphere, naturally that would affect the results. Everybody seems to agree there is a stratospheric contribution, even Spencer and Christy, the only difference is on how to correct for it. It seems like the many papers published in the last 5 years have come to an agreement, with Spencer and Christy in a small minority that disagrees. But then how do they differentiate between different channels that are meant to show the temperature at different levels of the atmosphere? And again, even if independent analysis shows .2C/decade warming in the mid troposphere, that's not the same as the LT satellite anomalies that are published. Link to comment Share on other sites More sharing options...
skierinvermont Posted February 7, 2011 Share Posted February 7, 2011 Correct me if I'm wrong, but that is different than UAH, right? I believe UAH is much closer to the Ch. 5 data....which I assume is different than Ch. 2. Also, if you see my previous post, I'm confused at to why the mid-troposphere is being referenced, when RSS and UAH anomalies were both supposed to be LT temperatures... There are only 4 channels on MSU channels 1-4 and RSS uses channel 2. Channel 2 has a weighting function 4-7km above the surface depending on the incidence angle of measurement. It's simply a "mean" so it contains contributions from the LT, MT, upper troposphere, and stratosphere. I think the Channel 5 might be on the new satellites that were deployed in the last 10 years but the majority of UAH and RSS use channel 2 MSU data from at least pre-2000. Link to comment Share on other sites More sharing options...
tacoman25 Posted February 7, 2011 Share Posted February 7, 2011 There are only 4 channels on MSU channels 1-4 and RSS uses channel 2. Channel 2 has a weighting function 4-7km above the surface depending on the incidence angle of measurement. It's simply a "mean" so it contains contributions from the LT, MT, upper troposphere, and stratosphere. I think the Channel 5 might be on the new satellites that were deployed in the last 10 years but the majority of UAH and RSS use channel 2 MSU data from at least pre-2000. Yeah, I was going to say the current AMSU data goes from Channel 4 to Channel 14. Link to comment Share on other sites More sharing options...
skierinvermont Posted February 7, 2011 Share Posted February 7, 2011 But then how do they differentiate between different channels that are meant to show the temperature at different levels of the atmosphere? And again, even if independent analysis shows .2C/decade warming in the mid troposphere, that's not the same as the LT satellite anomalies that are published. I really don't know how the satellite measurements physically work ... I just know that Channel 2 which is used for the majority of UAH/RSS dataset has a mean elevation of 4-7km. I guess that is arguable whether it is LT or MT since the tropopause is at 10-15km, perhaps that is why some say mid and some say lower. And that is only a "mean." It contains contributions from the stratosphere, upper troposphere and near surface. Anywhere that the satellite "beam" passes through. Either way, UAH and RSS contain a stratospheric contribution that biases them cold. Link to comment Share on other sites More sharing options...
tacoman25 Posted February 7, 2011 Share Posted February 7, 2011 I really don't know how the satellite measurements physically work ... I just know that Channel 2 which is used for the majority of UAH/RSS dataset has a mean elevation of 4-7km. I guess that is arguable whether it is LT or MT since the tropopause is at 10-13km, perhaps that is why some say mid and some say lower. And that is only a "mean." It contains contributions from the stratosphere, upper troposphere and near surface. Anywhere that the satellite "beam" passes through. Either way, UAH and RSS contain a stratospheric contribution that biases them cold. But if all levels are determined by a "mean of the beam"...wouldn't the stratosphere also be "contaminated" by the troposphere/surface? Wouldn't all levels be equally questionable? Link to comment Share on other sites More sharing options...
skierinvermont Posted February 7, 2011 Share Posted February 7, 2011 But if all levels are determined by a "mean of the beam"...wouldn't the stratosphere also be "contaminated" by the troposphere/surface? Wouldn't all levels be equally questionable? I would think so yeah.. let me read Fu et al. again. For some reason different frequencies seem to target different levels but I don't think there is a way that a beam can isolate a particular level. Link to comment Share on other sites More sharing options...
BethesdaWX Posted February 7, 2011 Share Posted February 7, 2011 I really don't know how the satellite measurements physically work ... I just know that Channel 2 which is used for the majority of UAH/RSS dataset has a mean elevation of 4-7km. I guess that is arguable whether it is LT or MT since the tropopause is at 10-13km, perhaps that is why some say mid and some say lower. And that is only a "mean." It contains contributions from the stratosphere, upper troposphere and near surface. Anywhere that the satellite "beam" passes through. Yes but it cannot contaminate if its not measured, or taken into account. The satellite is designed to hone in upon a certain region of the atmosphere, while ignoring other regions as if they're invisible. We have presumed that there could be factors handicapping satellite measurement, however, there are handicaps effecting every measurement system regardless. We cannot adjust for something that has no counter-verification. Link to comment Share on other sites More sharing options...
skierinvermont Posted February 7, 2011 Share Posted February 7, 2011 But if all levels are determined by a "mean of the beam"...wouldn't the stratosphere also be "contaminated" by the troposphere/surface? Wouldn't all levels be equally questionable? Fu et al. has a good image of the relative weighting functions of Channel 2 and Channel 4 on page 2 Figure 1. Channel 4 does have a little bit of troposphere contribution, but it's focused pretty high up in the stratosphere. http://www.ncdc.noaa...2524-UW-MSU.pdf Interestingly, when Fu et al. was published in 2004 UAH still had a decadal trend of .01K/decade, while RSS was .1K/decade. Both have been revised up significantly since then. Link to comment Share on other sites More sharing options...
BethesdaWX Posted February 7, 2011 Share Posted February 7, 2011 The "beam" of the satellite goes through the entire regions of the atmosphere, so "contamination" from one level to the next can be put in either "direction of influence", so to speak. Link to comment Share on other sites More sharing options...
tacoman25 Posted February 7, 2011 Share Posted February 7, 2011 Fu et al. has a good image of the relative weighting functions of Channel 2 and Channel 4 on page 2 Figure 1. Channel 4 does have a little bit of troposphere contribution, but it's focused pretty high up in the stratosphere. http://www.ncdc.noaa...2524-UW-MSU.pdf Interestingly, when Fu et al. was published in 2004 UAH still had a decadal trend of .01K/decade, while RSS was .1K/decade. Both have been revised up significantly since then. Right, I knew there was a major revision (I thought in the early 2000s) that pushed the satellite sources up quite a bit. This was due to a calculation error that Spencer/Christy eventually acknowledge and accepted. In fact, this might have been one of the main reasons RSS was created? But if that is in fact the case, then the current UAH/RSS anomalies probably take into account more of what Fu was referring to, since their trends are significantly warmer now than when he wrote that...and it obviously hasn't been because of a sudden/extreme warming. Link to comment Share on other sites More sharing options...
skierinvermont Posted February 7, 2011 Share Posted February 7, 2011 OK it appears RSS does try to subtract out the stratospheric contribution. Not sure whether UAH does. The following is the relative weighting profile for RSS. The graph that we are all familiar with is of the TLT data. I guess TLT is not just raw MSU CH2 or AMSU CH5.. it must subtract out the stratosphere. It does contain a significant contribution from the upper troposphere and maybe a little stratosphere as well. I guess one of the revisions in RSS incorporated Fu et al. 2004 or did something else to focus on the LT. Because at the time of Fu et al 2004, RSS only had a trend of .1K/decade and it's been revised up to .16K/decade. Link to comment Share on other sites More sharing options...
tacoman25 Posted February 7, 2011 Share Posted February 7, 2011 OK it appears RSS does try to subtract out the stratospheric contribution. Not sure whether UAH does. The following is the relative weighting profile for RSS. The graph that we are all familiar with is of the TLT data. I guess TLT is not just raw MSU CH2 or AMSU CH5.. it must subtract out the stratosphere. It does contain a significant contribution from the upper troposphere and maybe a little stratosphere as well. I guess one of the revisions in RSS incorporated Fu et al. 2004 or did something else to focus on the LT. Because at the time of Fu et al 2004, RSS only had a trend of .1K/decade and it's been revised up to .16K/decade. Ok, this makes a lot more sense. I know I've seen that graph before. And I'm sure UAH also makes an effort to subtract out the stratospheric influence (why wouldn't they?). That is all part of the analysis that both RSS and UAH do, and why they don't instantly have anomalies at the end of every month, I would guess...it's not just raw data they are publishing. If the theory about the revision having to do with incorporating Fu's analysis, wouldn't that argue against the warmer .2C/decade trend, since the satellite analysis is already taking stratospheric influence into account? Link to comment Share on other sites More sharing options...
skierinvermont Posted February 8, 2011 Share Posted February 8, 2011 Ok, this makes a lot more sense. I know I've seen that graph before. And I'm sure UAH also makes an effort to subtract out the stratospheric influence (why wouldn't they?). That is all part of the analysis that both RSS and UAH do, and why they don't instantly have anomalies at the end of every month, I would guess...it's not just raw data they are publishing. Yes it does make sense so your initial hunch was right. I just read though that RSS does not actually subtract out the stratosphere. They do something different that I don't understand to try and focus Ch2 (AMSU ch5) on the surface. I'll try to figure it out. It could be that the recent studies disagree with how UAH and RSS have gone about focusing on the LT. Regardless, it seems that the large majority of literature published in the last 3 years on both satellite data and radiosonde data finds a LT trend of .2C/decade. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.