Android

is that the north pole? in any case whats that interesting looking yellow thing?

Yes you are correct

Update. It's now about a year later. The main difference is that the recent El Nino is included. The solar minimum is still around! but the next cycle is clearly now ramping up. The parameters used for the graph below aren't exactly identical to the ones used above. It's still curve fitting so take it with a grain of salt anyway. Red: HadCRUT3 Orange: Sunspot count x0.0015 Maroon: MEI (Enso Index) x 0.12 and shifted forward by 110 days (this is arbitrary) Green: exponential background warming trend - cause unspecified - 0.0035*X^1.4 Light Grey: Volcanic forcing x 0.05 (this is merely to reduce the very obvious discrepancy around the time of the Pinatubo eruption) Blue line: Orange + Maroon + Green + Gray Everything is done by eyeballing a fit. No statistical procedure to determine scalings and offsets was used. I know the image quality, captioning, etc isn't that great either. But the analysis doesn't really deserve more than this. It's just rough curve fitting by eye, somewhat interesting but probably wrong in a number of ways. It's not even clear what all the assumptions are. Below is the same graph except the blue line does not include the sunspot cycle. This is to emphasize how factoring in the solar cycle seems to yeild a better match. I posted something similar on the thread a year ago showing that when I remove ENSO and the background warming trend there does appear to be a significant 11 year cycle in the data. Again eyeballing the data, so take it with a pinch of salt. Back to the first graph: If this is anything to go on (and it likely is not so this is just speculation), it is the recent drop in the solar cycle and the long sustained solar minimum that is responsible for the global temperature plateau/fall in the past ~7 years. Previous descending tail ends of solar cycles also appear to result in such falls in global temperature. The recent plateau appears more obvious because it is not as heavily distorted by volcanic and ENSO variability, and the solar minimum has been drawn out for longer. If the next solar cycle is similar to the last 3, the ramp up into solar cycle 24 (you can just see this starting at the end of the orange SSN plot) should yeild about 0.15C warming, which coupled with the ever rising green background trendline, will effectively bring global temperature to a new plateau. Ie global temperature will experience an upward step change some time within the next 3 years. I emphasize this is what will happen in to the blue plot, not necessarily to the red (hadcrut) one. It's possible the whole basis of this eyeballing is wrong and therefore that the blue and red plots will diverge in coming years.

interesting, plus the stuff on COWL in that paper.

Global map of SST anomalies as of last thursday: http://www.osdpd.noaa.gov/PSB/EPS/SST/data...ht.5.7.2009.gif You can compare for other dates here: http://www.osdpd.noaa.gov/PSB/EPS/SST/climo.html They also get updated twice a week (thursdays and mondays)

yes, on solarcycle24. Since the last post I have added in the models forcings you used for greenhouse gases, volcanic forcing, aerosols as well as extending it over the entire 20th century. I had to drop use of MEI as the ENSO index because it only goes back to 1950. So I am now using the global-sst enso index (http://jisao.washington.edu/data/globalsst...#digital_values). It's close to MEI anyway. The scales I have chosen are all arbitary and were chosen soley in attempt to get the best alignment with hadcrut. So this only suggests these scales are consistent with the temperature record, not that they are the actual case. This goes especially for the aerosol and greenhouse gas forcing scales. Because the aerosol forcing record and greenhouse gas forcing record oppose each other, there are probably a variety of different scales that can be used to get a match. The fit is fairly good: Note that the axis of the two graphs unfortunatly don't line up due to a pasting error. There is a small mismatch in the 40s. Another possible flaw is that unlike the leaky integrator on the other thread, there are no delays in response to forcing in this method (apart from the response to ENSO) I also performed a similar exercise to remove the volcanic, ENSO and solar cycle from the UAH record. It's a very rough "correction". But it does show that the recent flattening of temperature (since 2002) is consistant with the declining ENSO and solar cycle since then:

That is interesting, more detailed than my attempt. I avoided putting in greenhouse gases because I knew that would requrie aerosol forcing too, which I couldn't find data for. Knowing the sum of both greenhouse gas and aerosol forcing is a wiggly line over time meant I just stuck to starting in 1975 where my assumption was that the forcing of both would be fairly much consistant. I could add both in now using the ModelE aerosol forcing, although you have essentially already done that. I'll take a look at SOI too when I also take a look at AMO. Something bothers me about using these ocean oscillations and deducting them from global temperature, given that they are part of global temperature in the first place. Yes you might find a solar cycle in there when you deduct everything else from the temperature record, but I imagine the uncertainty might hide it. It's small and hard to see but it seemeed to make everything line up better, at least in my attempt since 1975. It would be interesting to extend the graph to make a prediction for the next two decades. I couldn't do that because my underlying trend was assumed. But assuming aerosl forcing, greenhouse and solar cycle can just be extrapolated it could be possible to make a prediction.

Good point I will look at AMO. Another thing I just did was to remove the MEI and the long term trend (using the out-of-thin-air 'exponential trend' used earlier of 0.0035*X^1.4) from hadcrut to see what is left to describe. The black line is the remaining temperature variation once ENSO (+PDO?) and this long term trend are removed. The pinatubo cooling is more obvious now (I assume that's what it is). One thing I did was plot sunspot number (scaled down a lot) from http://solarscience.msfc.nasa.gov/greenwch/spot_num.txt against this, which seems to line up fairly well. If this is anything to go on (and there are far too many assumptions to get to this stage) it suggests the solar cycle causes about a 0.2C change in global temperature from minimum to maximum.

I was going to adjust for PDO next to see how that affected things. There is suggestion that positive PDO over the past 30 years caused a significant part of global warming, and also that it's played a large role in recent cooling. Here's a graph of ENSO MEI index vs PDO index since 1950. The PDO record is so similar to the ENSO record that I seem to have already adjusted for it when I adjusted for MEI. I got the PDO numbers from http://jisao.washington.edu/pdo/PDO.latest That page says If I understand that correctly, PDO can be described as: Positive PDO means North pacific sea surface is warmer than global sea surface on average. Negative PDO means North pacific sea surface is cooler than global sea surface on average. Seems more simple than MEI which contains more factors than just sea surface temperature as VillagePlank mentioned. Given the proximity of North Pacific Ocean poleward of 20N to ENSO regions wouldn't it be the case that a strong El Nino would result in a warmer North Pacific Ocean sea surface temperatures? So could it be that PDO tends to follow ENSO because the North Pacific sea surface temperatures are influenced by ENSO events? The 1998 El Nino for example also saw PDO go strongly positive. PDO also went strongly negative during the 1999-2000 La Nina and during the recent La Nina. That seems too simple to me, but in any case if PDO and ENSO track so closely then there doesn't seem to be any reason to invoke PDO as a cause of climate change when this is little different from changes in El Nino/La Nina ratios and this is already provide by MEI. The question of PDO contribution to global temperature could therefore be phrased as "In the last 30 years there have been more El Ninos than La Ninas compared to the previous 30 years, what effect has that had on temperature?" Which I suppose can be done by removing the effects of ENSO from hadcrut temperarture record. Here is Hadcrut3 minus MEI (ie Hadcrut3 with ENSO adjusted out...somewhat..noway near perfect, it's a month for month subtraction, but it cannot be wildly wrong and the 1998 el nino is removed almost entirely) If this is anything to go on (perhaps not - it assumes PDO is driven by ENSO among other assumptions), it seems PDO has not contributed in any significant way to the recent warming trend. ENSO has only contributed to variation in global temperature and also not in any significant way to the recent warming trend. Added: I just remembered a better attempt at removing ENSO from hadcrut: http://www.realclimate.org/index.php/archi...rends-and-enso/

I know curve fitting is a bad thing, but I wanted to do it anyway. In red is hadcrut3: In blue added Multivariate ENSO Index (MEI) (http://www.cdc.noaa.gov/people/klaus.wolter/MEI/) Scaled the MEI down: Add a 0.15C/decade trend: This is the real fudge one, to get a better match I made the trend lightly exponential and also shifted the MEI plot forward by 110 days: In my opinion this suggests that the flat temperatures since 2002 are actually consistent with an accelerating global temperature trend in recent years that has been somewhat masked by the recent ENSO decline (similar to how pinatubo masked an otherwise warming trend). There doesn't seem to be any divergance between MEI and global temperature since 2002 like there was during the Pinatubo eruption period. There doesn't seem to be any obvious impact from solar minimum or from the negative PDO although of course either of those two could have caused the decline in MEI. In any case if the relationship over previous decades is anything to go by it would take progressively stronger La Ninas at this point to stall temperature any futher, which seems unfeasible. So when we are already low and there's not much further to go, my guess is that it's more likely than not that we will shortly see a sudden and large step jump like we did from 2000 to 2002.

Setting any lag time aside, remember that you need an increase in urbanization to cause an increase in warming. Just as you need the sun to increace in output or greenhouse gases to increase in concentration for an increase in warming. ie if UHI stays the same it won't result in a warming trend. If London, or rather the area of london around a temperature sensor, has had about the same level of "urbanness" for 30 years, which is probably true, then in that time period UHI shouldn't have contributed much if any to any increase in warming. Chinese cities on the otherhand have been built up recently and so they have gone from low UHI to high UHI in a short space of time that will cause a larger increase in temperature at those locations.

I don't know if the positive westerlies have held but SST in the east tropical pacific has increased significantly in the past week http://www.osdpd.noaa.gov/PSB/EPS/SST/data...t.3.26.2009.gif vs http://www.osdpd.noaa.gov/PSB/EPS/SST/data...ht.4.2.2009.gif http://www.cpc.noaa.gov/products/intraseas...850u_tlon.shtml http://www.cpc.noaa.gov/products/intraseas...heat_tlon.shtml

Easterly anomolies in the central pacific tropics just weakened and are almost negative (ie positive westerlies)..first time since september

He's right that the adverts are nonsense and a waste of time. He's wrong about the science though, his position is extremely illogical.

Isn't a comparison of (61-90) and (71-00) effectively a comparison of a single decade of change though? The GISTEMP anomaly map for this comparison backs up your numbers, a large part of spain has negliable warming: http://data.giss.nasa.gov/cgi-bin/gistemp/...200&pol=reg