Changes in atmospheric circulation and the Arctic Oscillation preserved within a millennial length reconstruction of summer cloud cover from northern Fennoscandia

Cloud cover currently represents the single greatest source of uncertainty in General Circulation Models. Stable carbon isotope ratios (δ¹³C) from tree-rings, in areas of low moisture stress, are likely to be primarily controlled by photosynthetically active radiation (PAR), and therefore should provide a proxy record for cloud cover or sunshine; indeed this association has previously been demonstrated experimentally for Scots pine in Fennoscandia, with sunlight explaining ca 90% of the variance in photosynthesis and temperature only ca 4%. We present a statistically verifiable 1011-year reconstruction of cloud cover from a well replicated, annually-resolved δ¹³C record from Forfjord in coastal northwestern Norway. This reconstruction exhibits considerable variability in cloud cover over the past millennium, including extended sunny periods during the cool seventeenth and eighteenth centuries and warm cloudy periods during the eleventh, early fifteenth and twentieth centuries. We find that while a generally positive relationship persists between sunshine and temperature at high-frequency, at lower (multi-decadal) frequencies the relationship is more often a negative one, with cool periods being sunny (most notably the Little Ice Age period from 1600 to 1750 CE) and warm periods more cloudy (e.g. the mediaeval and the twentieth century). We conclude that these long-term changes may be caused by changes in the dominant circulation mode, likely to be associated with the Arctic Oscillation. There is also strong circumstantial evidence that prolonged periods of high summer cloud cover, with low PAR and probably high precipitation, may be in part responsible for major European famines caused by crop failures.