Air/Stream Temperature Relationship Models And Stream Temperature Change In Paradise Watershed,Northeast Of Pennsylvania,U.S.

Temperature has significant impact on aquatic organisms. Stream temperature influences ecological processes of aquatic species such as growth, physiology, distribution, and plays the significant role in water quality. Historically, the high quality water, mountainous forests and low temperatures in Paradise watershed was ideal habitats for cold-water fish. However, recent urbanization, population growth and economic development impose a potential threat on degradation of the high quality habitats for cold water fish. This study focuses on air/stream temperature relationship models and whether the stream water temperature increases in Paradise Watershed. The main methodology is shown in the below: we set 35 monitoring sites in Paradise Watershed, Northeastern Pennsylvania. In order to know how human activities impact stream water temperature changes, we used HOBO pro temperature loggers to measure stream temperature of these 35 sites that above and below lakes, ponds, dams, and sewer/waste water treatment plant(SWWTP). The above ones were called control sties, whereas the below ones called test sites. The HOBO pro temperature loggers recorded temperature every 15 minutes at each site with approximately one-year period spanning from January 2009 to March 2010. The work and conclusion of this thesis is summarized as the following:(1) Air/stream temperature models. Air and stream temperature models were used to identify how climatic factor is influencing stream temperature. This research adopted linear and non-linear models to simulate the relationship between them, of which, non-linear models included Fourier model and Gaussian model. We found that the correlation between air temperature and stream temperature improved for each model as air temperatures were averaged over longer periods of time. Linear regressions showed the weakest correlation on average(0.8687, 0.9330, and 0.9504 for daily, weekly, and monthly time periods,respectively) and the Fourier function showed the strongest correlation on average(0.9339, 0.9721, and0.9781 for daily, weekly, and monthly time periods, respectively). The average RMSE decreased as time periods were extended as well. The linear regression RMSE's were better than both nonlinear regressions.The Fourier model had better RMSE's than the Gaussian model. The average coefficient determinant(R2)in test sites were lower than those in control sites on the daily, weekly, and monthly time scales. Test sitesare more weakly correlated from upstream sites due to impacts from SWWTPs or lake outflows. On the three time scales, the slope of linear model for two sites FOHI05 and YANK01 were smaller than 0.36, the rests had larger slopes. Compare the slopes of linear model in test sites and control sites, most of slopes in test sites were not declining, but increasing, which violated our expectation. In addition, most of sites(e.g.,SWIF04, FOHI02, and FOHI01) can observe the hysteresis phenomenon, which is an anticlockwise pattern over time of the air temperature and stream temperature relationship when plotted weekly or monthly.(2)The standard of Cold Water Fishery(CWF) and monthly stream water temperature violation percentage in Paradise Watershed. The temperature readings were compared with CWF that was defined by Environmental Protection Agency of Pennsylvania to calculate the percentage of violations that occurred at each site. The streams with high violation may have the high probity of being thermal degraded. From 1991 to 2010, the correlation between air temperature and violation percentage in summer and winter are positive, but weak, the correlations were 0.2385 and 0.0901, respectively. We found that the high volition often occurred in March, most of test sites had higher violation in winter(e.g. SWIF04 and SWIF06), while some of other sites had higher violation percentage in summer(e.g.FOHI02). Swiftwater Creek and Forest Hills were the most influential tributaries that the thermal degradation occurred. In spite that this research confirmed that dams, lakes, ponds, and SWWTPs are the main factor to result in water temperature violations, we cannot conclude that the stream temperature was indeed increasing in Paradise Watershed due to the cross-sectional nature of the analysis.(3)The patterns of maximum stream temperature change in a longitudinal gradient in Paradise Watershed. The main stem of Paradise Creek's maximum stream temperature follows a pattern of increasing stream temperature on a longitudinal gradient downstream for its' entire length, therefore, it is likely for maximum steam water temperature in downstream to excel the thermal limit of brook trout,especially in July and August of 2009(i.e. FOHI02). However, due to the effects of stream buffers and insulators in riparian zone, the temperature in the most downstream sites(i.e. OHI01), on the contrary, was decreasing under the thermal limit of brook trout. Analysis of daily maximum stream temperature on tributaries in Paradise Watershed indicates that stream buffers and insulators play a vital role in reducing the thermal degradation.This result of this research indicated that the tributaries in Paradise Watershed has being exposed the thermal degradation. The degradation came from multiple sources, such as dams, ponds, lakes, and SWWTPs, are major factors to cause the high violation percentages. The low coefficient determinant of air/stream models, the high number of readings are above the CWF standard, as well as the high violation percentages, which all indicates that the thermal degradation has been occurred in the watershed. It is necessary to take actions to reduce or remove the input of these thermal resources. The practical and effective keys to dealing with the thermal degradation is the maintenance of the insulating and buffering processes(tree plantings or bank stability projects) to resist warmer air temperatures that are projected to increase in Pennsylvania. Continued monitoring and improvement of techniques to control stream temperature will be necessary to combat human disturbances that may increase stream temperature in the future in the Paradise Creek Watershed.