Climatic Change

, Volume 111, Issue 2, pp 335–363

The earliest temperature observations in the world: the Medici Network (1654–1670)


    • National Research Council (CNR), Institute of Atmospheric Sciences and Climate
  • Chiara Bertolin
    • National Research Council (CNR), Institute of Atmospheric Sciences and Climate

DOI: 10.1007/s10584-011-0142-5

Cite this article as:
Camuffo, D. & Bertolin, C. Climatic Change (2012) 111: 335. doi:10.1007/s10584-011-0142-5


This paper presents the earliest temperature observations, scheduled every 3–4 h in the 1654–1670 period, which have been recovered and analysed for the first time. The observations belong to the Medici Network, the first international network of meteorological observations, based on eleven stations, the two main ones being Florence and Vallombrosa, Italy. All observations were made with identical thermometers and operational methodology, including outdoor exposure in the shade and in the sunshine to evaluate solar heating, state of the sky, wind direction and precipitation frequency. This paper will consider only the regular temperature series taken in the shade. The observations were made with the newly invented spirit-in-glass thermometer, also known as Little Florentine Thermometer (LFT). The readings have been transformed into modern units of temperature (°C) and time (TMEC). The LFT has been analysed in detail: how it was made, its linearity, calibration and performances. Since the middle of the LIA, the climate in Florence has shown less than 0.18°C warming. However, although the yearly average showed little change, the seasonal departures are greater, i.e. warmer summers, colder winters and unstable mid seasons. The temperature in the Vallombrosa mountain station, 1,000 m a.m.s.l, apparently rose more, i.e. 1.41°C. A discussion is made on the interpretation of this finding: how much it is affected by climate change or bias. A continuous swinging of the temperature was observed in the Mediterranean area, as documented by the long instrumental observations over the 1654–2009 period. However, changes in vegetation, or exposure bias might have contributed to reduce the homogeneity of the series over the centuries.

Supplementary material

10584_2011_142_MOESM1_ESM.pdf (4.2 mb)
ESM 1(PDF 4281 kb)

Copyright information

© Springer Science+Business Media B.V. 2011