Climate change in nontraditional data sets.

Scientific examinations of nontraditional data sets collected by naturalists, hobbyists, or indigenous peoples have been instrumental in linking changes in biological communities to recent climatic changes. Many phenological records in particular show close correlations between timing of biological events and climatic conditions (1). The timing of these biological events is likely to be related to climate warming through intermediary processes that affect food and habitat availability, such as snowmelt or ice thaw. Here we report an unusual 84-year data set on timing of river ice breakup that shows temporal patterns consistent with previously reported trends toward earlier spring. The Nenana Ice Classic is a betting contest held in Nenana, Alaska (64°34!N, 149°05!W), in which participants attempt to guess the exact minute in spring that a wooden tripod positioned on the frozen Tenana River will fall through the breaking ice (Fig. 1). Started as a diversion for railroad engineers in 1917 with a pot of $800, the contest has grown into a popular event in which hundreds of thousands of entrants compete for jackpots exceeding $300,000. Contest records of the exact minute of ice breakup date back to 1917 and can be considered quite accurate as the high stakes lead to constant vigilance at the time of breakup. The tripod location was moved away from a railroad bridge before the 1970s (when heavier and more frequent train traffic began) and is occasionally moved several meters away from gravel bars but remains in a straight section of the river that has not altered course substantially (2). River ice breakup is known to be due to a combination of thermal effects (direct melting) and dynamic effects (mechanical forces from upstream drift ice) (3). Warmer climate would be expected to advance the time of breakup through both thermal effects and dynamic effects, due to thinning ice and increased snowmelt runoff into rivers (3). River ice breakup has direct economic consequences, as breakup marks the beginning of waterborne commerce and travel between high-latitude population centers. Ecologically, ice breakup is a time of rapid temperature increase and nutrient loading (3). Changes in the timing of ice breakup thus would be expected to have “bottom-up” ecosystem effects, although these have not been documented. We examined the entire record of ice breakup (4) to look for long-term trends in breakup. We compared the ice breakup record from 1949–2000 with available climatic data for Nenana and Fairbanks, Alaska (90 km away). Climatic data for January to April (the months preceding ice breakup) of each year obtained from the Western Regional Climate Center (5) included daily precipitation (PCPN), snowfall (SNFL), and air temperature minima (TMIN) and maxima (TMAX). Years in which any month had fewer than 15 days of data were removed from analysis, leaving 32 to 35 years of data for the Nenana records and 47 to 51 years for the Fairbanks records. Heat-island effects due to urbanization and movement of the weather station may affect the Fairbanks record. By con-