390 



Global Climate Clums^e — Our Living Resources 



Climate 

 Change in the 

 Northeast 



by 



Ray T. Oglesby 



Charles R. Smith 



Cornell University 



Fig. 1. New York locations from 

 which phenological data were 

 obtained. I -Allegheny Plateau 

 (birds); 2-Cayuga Lake basin, 

 Ithaca (birds); and 3-Hudson 

 Highlands (flowering plants and 

 birds). 



Fig. 2. Trend for hepatica 

 {Hepatica aculiloha) from the 

 Hudson Highlands (Mohonk 

 Preserve) of southeastern New 

 York, showing tendency for earlier 

 spring blooming. The negative 

 slope of the trend is significant at 

 P < 0.05. 



Climate is a principal lieterminant of biolog- 

 ical (distributions and of patterns that char- 

 acterize the seasonal physiology and behavior 

 of many oiganisms (Gates 1993). Consequently, 

 a changing climate should elicit responses in 

 these biological properties. Detecting and char- 

 acterizing such changes are logical early steps 

 in assessing the significance of climate change 

 to species and ecosystems (Schwartz 1990). 

 Most published work on this subject involves 

 species and ecosystem modeling based on 

 known physiological and behavioral traits of 

 selected species. This article presents evidence 

 froiTi an aiTay of phenological data suggesting 

 that climate change is occuiring and that its bio- 

 logical effects may already be of considerable 

 magnitude. (Phenological data are those associ- 

 ated with the relationship between climate and 

 periodic phenomena like bird migration and 

 flowering.) 



Most research intended to explore possible 

 effects of climate change on vegetation has 

 understandably focused on agricultural and for- 

 est plants. Our approach, however, focuses on 

 examining historical trends at the regional level 

 and identifying species of potential value as cli- 

 mate change indicators. With the assumption, 

 based on climate niodels. that unidirectional 

 warming is already occurring and will probably 

 accelerate over the next few decades, we began 

 to search for evidence of biological responses 

 among very different groups of organisms. 



May 30 ■ 



Hepatica 



r=678-0.29x 

 fl^ = 13% 



Apr 30 - 



ra Mar 31 



Mar 01 



"TT- 

 25 



39 



I 1 n M M I I 



53 67 



Year 



I I I I I M 



81 



Specifically, dates of the first return of spring- 

 migrating birds and of the first bloom of spring 

 wildflowers in the Northeast were sought in 

 long-term (50 or more years), continuous, reli- 

 able records. 



We computerized and analyzed two major 

 and several minor long-term data sets from 

 handwritten lecords from three New York State 

 locations (Fig. 1 ). An especially rich source was 

 records from the Cayuga Biid Club at Ithaca. 

 Highly reliable observers recorded first spring 

 sightings of migratory birds from 1903 to the 

 present in the Cayuga Lake basin of central 

 New York as delineated by Wiegand and Eames 

 (1925). A second source of extensive, high- 

 quality information was records for dates of 

 first spring airival for migrating birds and dates 

 of first bloom for spring wildflowers at the 

 Mohonk Preserve, an upland site in the mid- 

 Hudson Highlands region of southeastern New 

 York; these records extend from the late I920's 

 onward. Both sites are expected to continue 

 generating comparable data sets. A third data 

 set includes dates of first spring arrival for 

 Louisiana waterthrush (Seiunis motacilla) and 

 solitary vireo (Vireo solitarius) in western New 

 York (1952 to present) on the Allegheny 

 Plateau. 



Our general approach to data analysis has 

 been in the form of species plots with date of 

 first arrival or first bloom as the vertical axis 

 and sequence of years as the horizontal axis 

 (Figs. 2-4). 



Status and Trends 



Flowering Plants 



Phenological data were examined for 15 

 species of spring wildflowers on time of first 

 blooming at the Hudson Highlands site (Fig. 2). 

 Six species of wildflowers all exhibited signifi- 

 cantly earlier (P < 0.05) rather than later bloom- 

 ing (averaging -19.8 days/50 yr: /?- = 0.26). The 

 remaining nine species showed no significant 

 patterns of change. We only can speculate why 

 six species exhibited such a pionounced change 

 and nine others did not. Clues may be obtained 

 when existing data for other plant species at this 

 site are examined. For example, the set of 

 species showing earlier blooming appears to 

 include plants typically found in more open 

 locations where soil temperature would show 

 the earliest and most rapid response to warm- 

 ing. One woody shrub, common witch-hazel 

 (Hciinanielis virginiana). which blooms in early 

 fall, also showed a significant trend toward ear- 

 lier bloom. 



95 



