Stream ecosystem responses to the 2007 spring freeze in the southeastern United States: unexpected effects of climate change

Abstract

Some expected changes in climate resulting from human greenhouse gas emissions are clear and well documented, but others may be harder to predict because they involve extreme weather events or heretofore unusual combinations of weather patterns. One recent example of unusual weather that may become more frequent with climate change occurred in early spring 2007 when a large Arctic air mass moved into the eastern United States following a very warm late winter. In this paper, we document effects of this freeze event on Walker Branch, a well-studied stream ecosystem in eastern Tennessee. The 2007 spring freeze killed newly grown leaf tissues in the forest canopy, dramatically increasing the amount of light reaching the stream. Light levels at the stream surface were sustained at levels considerably above those normal for the late spring and summer months due to the incomplete recovery of canopy leaf area. Increased light levels caused a cascade of ecological effects in the stream beginning with considerably higher (twothree times) rates of gross primary production (GPP) during the late spring and summer months when normally low light levels severely limit stream GPP. Higher rates of stream GPP in turn resulted in higher rates of nitrate (NO 3 À ) uptake by the autotrophic community and lower NO 3 À concentrations in stream water. Higher rates of stream GPP in summer also resulted in higher growth rates of a dominant herbivore, the snail Elimia clavaeformis. Typically, during summer months net NO 3 À uptake and snail growth rates are zero to negative; however, in 2007 uptake and growth were maintained at moderate levels. These results show how changes in forest vegetation phenology can have dramatic effects on stream productivity at multiple trophic levels and on nutrient cycling as a result of tight coupling of forest and stream ecosystems. Thus, climate change-induced changes in canopy structure and phenology may lead to large effects on stream ecosystems in the future.

Agricultural land use alters the seasonality and magnitude of stream metabolism

“…GPP tended to be higher in streams draining row-crop agriculture in this study (range: 0.1-14.4 g O 2 m 22 d 21 ) than in less disturbed, primarily closed-canopy headwaters across North America (Mulholland et al 2001;Bernot et al 2010); however, rates were as high as in Sycamore Creek (15 g O 2 m 22 d 21 ), an open-canopy stream in the arid southwestern United States (Mulholland et al 2001). Further, the range of GPP reported in first-and secondorder streams in an agricultural watershed in Illinois (, 0.1-15.5 g O 2 m 22 d 21 ; Wiley et al 1990) was nearly identical to the range we report in agricultural streams in northwestern Indiana.…”

“…Heterotrophic respiration of algal exudates may also partially account for this tight coupling between GPP and ER. This correlation between GPP and ER is more commonly found in open-canopied streams (Wiley et al 1990;Bunn et al 1999) than in closed-canopied systems (Mulholland et al 2001) except during periods of high autotrophic production (Roberts et al 2007).…”

“…Further, dissolved nutrients were not correlated with ER in Shatto Creek, and temperature explained only 2% of the variation in ER. Although some studies have found an effect of temperature on ER (Sinsabaugh 1997;Yvon-Durocher et al 2012;Beaulieu et al 2013), other variables such as allochthonous inputs or increased nutrients can be a stronger driver (Mulholland et al 2001;Roberts et al 2007). In Shatto Creek, the temporal pattern in ER tended to mirror that of GPP, especially in the spring through autumn months (Fig.…”

“…Although the seasonal patterns in GPP differ in agricultural and forested streams, metrics of light availability and penetration (as PAR and/or turbidity) remain strong predictors of GPP in both systems. Variation in nutrient availability has also been shown to control stream GPP (Mulholland et al 2001); however, the consistently high concentrations of DIN typical of agricultural streams (David et al 1997;Royer et al 2004Royer et al , 2006 are likely in excess of demand, and thus may not limit autotrophic production in agricultural streams. In contrast, SRP concentrations are low relative to DIN in agricultural streams (range of DIN : SRP ratios: 642 : 1 to 2098 : 1), and may exert some control on rates of GPP in Shatto Creek.…”

“…Rates of GPP are influenced by the availability of light (Hill et al 1995;Mulholland et al 2001;Fellows et al 2006) and nutrients (Grimm and Fisher 1986;Lamberti and Steinman 1997;Hall and Tank 2003). ER encompasses both autotrophic and heterotrophic respiration, and rates of ER are influenced by nutrients (Mulholland et al 2001;Hoellein et al 2007), water temperature (Sinsabaugh 1997;Uehlinger et al 2000), and organic matter availability (Acuñ a et al 2007;Roberts et al 2007). Open-canopied streams draining agricultural land are assumed to be autotrophic (GPP .…”