We are all familiar with the usual metrics used to highlight that the climate is changing: surface air temperatures, sea level, sea surface temperatures and ocean heat content all rising, glaciers retreating, Arctic sea ice declining etc. But, there are also many other less well known sources of information about how our climate is changing, and many involve ‘citizen scientists’, who often didn’t realise the potential long-term benefits of the data they were collecting.
Without reliable meteorological observations over the past few centuries (or access to a time machine), climate scientists have to rely on piecing together evidence of historical changes in climate from a wide range of sources. Interpreting this data is often not straightforward, but a consistent picture does emerge.
For example, there has been a recent study examining ice freeze dates in a lake in Japan back to 1443, and ice break-up dates in a river in Finland back to 1693 (also summarised by National Geographic). The Japanese measurements were taken by monks as part of a religious ceremony, and the Finnish measurements were started by a merchant involved in river trade. Both show changes over time, consistent with warming temperatures, although local non-climate factors will also clearly be important (Figure 1).
Similarly, in North America, the annual dates in the spring when winter ice cover leaves a lake has come to be known as ‘ice out’, and these dates have been measured by those who live on the lakes since 1836. A recent study analysed records from 12 lakes (Figure 2) and all show consistent trends towards earlier ‘ice-out’ dates. There is also considerable variability, likely due to contributions from variations in the atmospheric circulation, such as the North Atlantic Oscillation.
Another source of not-often-discussed climate data is lake temperatures. Summer lake temperatures, using in-situ measurements and satellite data, have increased in the vast majority of lakes across the world (Figure 3). These changes are important as warming surface waters means less mixing with deeper waters, depriving the lower layers of oxygen, with harmful consequences for lake marine life. Longer records based on only in-situ data back to 1910 are also under construction.
Moving to the natural world, one of the longest records is from Japan, where the flowering of Cherry trees has been recorded for centuries. A study in 2009 highlighted considerable variability in the flowering day in Kyoto, with a recent trend toward earlier dates (Figure 4). A temperature increase of 1°C moves the flowering dates around 3-5 days earlier. Urbanisation effects are important in this example, but trees from a range of local environments allows the climate related signal to emerge.
A final recent example is grape harvest dates in the Czech Republic (available since 1499), which have become much earlier in recent decades due to rising temperatures (Figure 5). There are obvious non-climatic factors which influence the date of grape harvest (e.g. wars, taste etc), but the grape type has apparently been rather constant.
Overall, these non-traditional measurements of climate tell a consistent picture of a warming world, with considerable interannual and decadal variability, and local non-climatic factors affecting many aspects of traditional events and measurements.
More needs to be done to rescue the billions of instrumental weather observations taken since ~1800, which are currently languishing in various archives around the world (e.g. OldWeather.org), but I would be interested to hear of any ‘data rescue’ activities needed for these non-standard types of observation too.