In this paper homogenisation methods are blindly tested on a dataset with unprecedented realism. All the most used and best algorithms have participated.

The two main conclusions are.
1. Homogenisation improves climate data. Because the test was blind and because of the realism of the data, this can now be stated with confidence.
2. Modern algorithms, which are designed to also work with an inhomogeneous reference, are clearly better than traditional ones. It needed a realistic benchmark dataset with surrogate climate networks to see this difference clearly.

A longer introduction to this paper can be found on the blog of Victor Venema:
<http://variable-variability.blogspot.com/2012/01/new-article-benchmarking-homogenization.html>

A direct link to the (open access) article is:
<http://www.clim-past.net/8/89/2012>

The benchmark dataset is kept available for people who would like to test their algorithms (in a non-blind fashion):
<http://www2.meteo.uni-bonn.de/venema/themes/homogenisation/monthly_homogenization_HOME.html>

(Updates going from the older to newer ones.)

v8.514
- bug in 1-1 repaired (warning about wrong multi ID settings)
v8.517
- bug repaired for 8-4, not all results from AnClim could be taken
v8.519
- 2-10  – made case insensitive
- 1-2 – more options for columns renaming
- show_dbf – opening in Excel sorted
v8.520
- new way of update – self-extracting zip (exe) file
- 2-4, differences, or put data aside or put data after
v8.523
- minimu mumber for map calculation (4-12)
v8.525
- 1-1 function (daily mode) repaired  (converting to months in column and empty info_file)
v8.526
- 1-8 function repaired
v8.526
- 2-1 repaired
v8.531
- messagebox (messages) – always on top (all the windows)
- launching R functions improved
- 1-10 (number of stations) function improved
v8.533
- registry reding for Windows 7 repaired
- 10-4 adapted also for days in column data structucture
v8.534
- 3-7 for days in columns quicker
v8.537
- 6-5 quicker for larga datasets,  appending another file improved
v8.539
- update from a file function included
v8.543
- creating benchmark dataset (from neighbours) for methods evaluation (2-22-6, 8-7)
v8.544
- cross table – sorting columns – repaired
- 9-2 and 9-3 repaired
- 10-2 monthly repaired
v8.551
- 10-2, 10-1 functions improved, correcitons with correlation below limit not accepted, new Inhom corr. file created without such cases
v8.556
- batchjob file launching from command line improved
- 2-8 and 2-6 improved for different elements in columns (HOBO outputs)
- 2-1: own scripts to be applied on input or output files, improved
v8.560
- 2-1: bug with header (column names) repaired
- 9-2: merging seasons included
- batchjob processing improved
- launching ArcView improved (re-launch when problem)
v8.564
- 2-11 function widened
v8.564
- 7-1 and 8-2 selection for dates repaired (in the middle of a year)
- 7-4: QC: according to IQR_time repaired,  o flag not applied any more (made RMSE worse)
v8.568
- 8-4: multi_ID accepted for “single” Info_file ID (combination with Region column)
v8.568
- 2-11 functionality added (info about files)
- 2-9  functionality added (commands upon input files possible, deleting rows before joining)
v8.573
- 8-5 for testing individual neighhours: improved
v8.575
- no Adjustment below correlation option: you can specify minimum number of values that has to be below the limit, to discard the adjustment for the particular year
v8.576
- 10-2 near inhomogeneities wrong deleting repaired, 9-6 in accordance
- appending with 0 instad of blank repaired
v8.577
- 9-3 repaired and improved (number of lines), 6-5: excluding dates outside limit – bug repaired
v8.584
- 5-5: RMSE and other statistics, better output,
- new Show_DBF version (in XII 2010, 2.6x) – tracking changes (through buferring)
v8.584
- 10-1 and similar: correlations are now from the 1st diffs. (it was not!!!),   warning about 1 dec. place for precip.
- batchjob: option for checking if the input files and options are correctv8.584
v8.589
- menu 1 adjusted to be the same (with items), both in monthly and daily mode ,   2-11 and 1-10 functiones improved
v8.592
- distributions (4-6, 4-7, 4-8 functions)> output adjustments,  statist menu: support for stations in columns
- 4-1 improved
- 3-1 transform: external average in the file possible
v8.601
- 2-16: distributes more columns into columns
v8.603
- 7-7: repaired bug with long gap (and not accepting the series)
v8.604
- 1-1: separating stations into more parts with gaps less than one year (for Days in Columns)
v8.605
- 6-1: first and last occurance: better description in the output
- 10-2: date_inhom column information added
- 5-9: SPI calculation improved
v8.608
- 6-5, k-minutes rainfall, output improved
- 1-1: filling gaps for less than one year – for days
v8.611
- multi select of columns
- import from Excel improved
v8.616
- 1-3: interpolating minute rainfalls between two records, 6-5: k-min. rainfalls improved
v8.617
- 1-2: applying scripts even for files imported from Excel
v8.624
- 3-11: frequencies: new options (individual years, relative frequencies)
- 2-2: saving into Excel improved (into individual sheets)
v8.632
- 1-1: converting to months in column problem repaired
- 7-7: reconstruction: improved (leting only part of the whole reconstructed length)
- 2-2: output into Excel improved – into sheets
- 10-1: correction using radios improved,
- 8-4: correlation used for excluding neighbours
v8.637
- 3-10 (differencing) improved, 4-1: sorting input according to time (for trend calculation)
- 1-5 (add geography) improved
- 2-1: importing from Excel: 2007-2010 MICROsoft Excel does not support saving into DBF IV … better not to comment, if you do not stay with “normal” (2003) version, now it is converted through CSV files …
v8.640
- 7-7 function improved
v8.643
- 8-5 repaired
- launching ArcView better controled
v8.647
- lables for Esc etc. improved
v8.647
- 4-1 bug fixed (daily mode, for days in column)
v8.647
- 1-1 (daily) and 8-5 bux fixed (exporting the test results)
v8.649
- 2-11 and 2-1  improved – subdirs option added
v8.658
- tuned problems in Win7
v8.662
- 5-2 function for daily data: median for given regions
v8.667
- COST data import and export functions implementation
v8.673
- waves, growing seasons estimation (6-1 and 6-5): problem with over year exceedance solved
v8.675
- 2-8: bug repaired
v8.679
- 6-5: k-days, k-min option repaired and improved
- 2-2: info_file now accepted (bug repaired)
v8.681
- automatic propagation of data file and data info file enabled (check option added)
v8.682
- 2-2: new feature added: possiblity to add a command or script to be processed upon the file before exporting (like 1col etc.)
- 2-10: function adopted to replace values in columns (like days) and define only upper part of a interval
v8.685
- 6-7 function (daily): extremity of rainfall intensities according to Wussov improved
- 3-13: getting grid points: improved
v8.685
- 6-7: plots can be created under R using context menu for the second output file