Homogenization of global early instrumental data

Instrumental meteorological observations are crucial for the analysis of climate backwards in time to reconstruct climate variations. However, the collection of instrumental data dating back to 1658 allows many of the long climate series to have often been affected by inhomogeneities (artificial shifts) due to changes in measurement conditions (relocations, instrumentation, change in environment, etc.). To deal with this problem, homogenization procedures have been developed to detect and adjust inhomogeneities. Homogenization in climate research means the removal of non-climatic changes. Next to changes in the climate itself, raw climate records also contain non-climatic jumps and changes for example due to relocations or changes in instrumentation.

This presentation describes the homogenization of the early instrumental dataset (HCLIM) of monthly mean temperature time series and other parameters such as precipitation and air pressure.

New homogenization algorithm validation methodology will be assessed by early instrumental data, and its use to assess the skills of three different algorithms, when applied to early instrumental data. The methods tested were PHA, HOMER and SPLIDHOM.