Comments 1 to 18:

1. robert way at 11:57 AM on 10 February, 2011
   Great resource, I highly recommend Kelly's site.
   0 0
2. Spencer Weart at 12:19 PM on 10 February, 2011
   There's a peer-reviewed paper that matches the temperature curve by combining volcanoes, El Nino, Sunspots, and anthropogenic emissions: Lean & Rind (2009), Geophysical Research Letters 36, L15708. See http://www.aip.org/history/climate/summary.htm#model
   0 0
3. Daniel Bailey at 13:15 PM on 10 February, 2011
   Thanks, Kelly! I visit your site daily during the Arctic melt season; weekly at this time of year. Highly recommended. The Yooper
   0 0
4. keithpickering at 15:21 PM on 10 February, 2011
   You should also include TSI, Total Solar Irradiance, in your time series suite. For recent values, PMOD observations are best. If you want to go farther back than 1978, the reconstruction of Lean et. al. can be found on Climate Explorer. The biggest problem with the Lean reconstruction is that it doesn't model the very deep decline seen at the end of cycle 23 very well. Which means that for recent values, PMOD observations are still better than any reconstruction. It is possible to create a fairly seamless composite dataset (PMOD 1978-present, Lean further back) by regressing Lean against PMOD during the overlap period 1978-2000, i.e., before cycle 23 reached its peak. You can then modify Lean's reconstruction by the regression slope and intercept (r>.96) and get a fairly unified monthly dataset going back to 1882.
   0 0
5. keithpickering at 15:24 PM on 10 February, 2011
   I should also mention: I've found that Nino4 has slightly higher correlations with global temps than Nino34 (as you're using) when best lags are used ... you might want to try both.
   0 0
6. Jeff Freymueller at 15:53 PM on 10 February, 2011
   Dealing with all sorts of different file formats is a problem for professional scientists, not just citizen scientists! There is nothing like providing code to read the files to help make analysis easier, so thanks for making yours available.
   0 0
7. hohygen at 19:58 PM on 10 February, 2011
   Great work. This will making climate communication easier. On comment on normalisation: From past experience with the temperaturedatasets: As far as I know do they use different normal period as reference. I cannot see any part of your text above commenting on this. My question is: Do you use the agencies normal period, og do you adjust the series to a common period?
   0 0
8. hohygen at 20:08 PM on 10 February, 2011
   A series of arctic ice coverage http://nsidc.org/arcticseaicenews/ (and ice volume: http://psc.apl.washington.edu/ArcticSeaiceVolume/IceVolume.php) would have been great.
   0 0
9. D Kelly O at 01:06 AM on 11 February, 2011
   hohygen "Do you use the agencies normal period, or do you adjust the series to a common period?" I use the agencies normal baseline period. I document this here.
   0 0
10. Jeff T at 03:02 AM on 11 February, 2011
    Kelly, thank you so much. There are many other data files that could be added, of course. My nomination would be sea-level data from U. Colorado.
    0 0
11. JMurphy at 03:29 AM on 11 February, 2011
    There's been so much so-called skeptical diversion and misinformation on here recently that it's a pleasure to see a post and an informative link from someone (Spencer Weart, above) who actually knows what they are talking about !
    0 0
12. D Kelly O at 04:09 AM on 11 February, 2011
    Jeff T Here is a link to a post on U of Colorado sea level trend ( link). I update the chart monthly. The post includes a link to the RClimate script for retrieving and plotting the U of Colorado NetCDF file. Kelly
    0 0
13. jyyh at 17:08 PM on 11 February, 2011
    Thank you for this compilation. If I use this how should I refer to this? Anyway I'm not planning to publish anything based on this in a serious scientific journal so the WUWT style of reference (i.e. none credible) might be enough?
    0 0
14. Kevin C at 01:14 AM on 13 February, 2011
    How about adding predicted sunspot numbers for the next cycle? Data here: http://solarscience.msfc.nasa.gov/predict.shtml http://solarscience.msfc.nasa.gov/images/ssn_predict.txt Given that the next maximum is currently looking to be very weak indeed (ssn of 58 vs 118 in 2000) this is very relevent to climate over the next 5-10 years.
    0 0
15. caerbannog at 06:00 AM on 13 February, 2011
    Thought that the citizen-science thread would be a good place to post this. Not enough material to justify a full guest post, so I'm just leaving a little note here. I've added some new features to my "Quick and Dirty" global temperature anomaly app. It now performs a simple gridded average (geospatial weighting) that produces results that are remarkably similar to NASA's "Meteorological Stations" temperature index (NASA results here: http://data.giss.nasa.gov/gistemp/graphs/ -- scroll down to the Global Temperature (meteorological stations) plot). The app now allows you to generate urban vs. rural results side-by-side (it reads the GHCN v2 metadata file to extract rural and urban stations separately). You can now debunk the "Urban Heat Island" talking-point in real-time. The all also allows you to generate ensembles of results, each computed from a different set of randomly-chosen temperature stations. You can throw out 90 percent of the temperature stations and still get results that are quite consistent with the results you get when you process all of the data. Here's a plot of the program's output vs. NASA's equivalent results (note: I used GHCN "raw" monthly-mean data): For those who are interested, the latest source-code can be found here: http://forum.signonsandiego.com/attachment.php?attachmentid=8142&d=1297529025 (Run the app without any arguments to get a semi-helpful "usage" message). Now, back onto the soapbox for a bit: It was surprisingly easy to implement the gridding/geospatial-weighting routine (much easier that I originally thought that it would be). The result posted above is what popped out on my first "full-up" run -- i.e. not tweaking or data-fiddling required. Only a few modest changes to my app would be required to perform all the temperature-data analysis/processing that Anthony Watts has been promising (but has failed to deliver) for years! Getting a crude version of this app up and running took something like a weekend in my spare time -- the bulk of the time spent was cleaning it up and getting it into a form where others might find it useful. The one major thing that I learned from this project is how breathtakingly inane and brain-dead the denialists' campaign against the surface temperature record has been. A major part of the "climategate" campaign against the CRU was based on the claim that nobody could verify the CRU's global temperature results because the CRU supposedly was "hiding" a small fraction of the data used in their global temperature calculations. That, of course, is completely absurd -- what my exercise demonstrated is that someone with basic programming skills and the ability to read documentation should have no trouble validating the CRU's results with just a weekend's worth of effort. There is enough public temperature data and plenty of free software tools to enable any technically competent person to debunk the entire basis of the so-called "climategate" scandal with just a couple of days worth of effort. The "climategate" campaign truly has taken "stupid" to unprecedented heights. I'm keeping my app handy so that the next time some loudmouth starts spouting off to me about NASA's/CRU's "cooked" data, I'll be able to whip out my laptop and shut him up. I've found that having someone see me generate results "on the fly" from my own app makes a bigger impression than handing him/her a link to a NASA web-page. Hopefully, others can find this app useful in the same way. I've given it a fair bit of testing, so I don't think that there are any "showstopper" bugs (but no guarantees!).
    0 0
    Moderator Response: [DB] Converted URL's to links.
16. Rob Painting at 06:53 AM on 13 February, 2011
    Caerbannog - I liken "skeptics" to dogs that chase after cars. They get all excited and noisy, but when the car stops they have no idea what to do. Same with skeptics and the raw data.
    0 0
17. Kevin C at 07:08 AM on 13 February, 2011
    Nice work Caerbannog! I had one interesting idea on a variant of the baselining technique: Calculate a baseline on the whole period rather than 1950-1980 (which will therefore be poor), then calculate the geographically averaged monthly anomaly. Then go back and recalculate the baseline for each station subtracting the monthly average to get a better set of baselines. Calculate the monthly anomaly again. Calculate baselines again. Iterate until stable. The result should be a poor-man's approximation to Nick Stokes' TempLS algorithm.
    0 0
18. makiem at 13:18 PM on 3 August, 2015

    Climate agencies provide a great service by putting their data files online, however, the organization and format of these files often makes it challenging for citizen scientists to compare series.

     

    0 0
    Moderator Response:

    [PS] Commercial link deleted. If this account is used for spam, then it and all associated posts will be deleted. One and only warning.

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.