AQUATOX輔以穩定同位素技術評估溪流生態受溫泉之影響

Abstract

溫泉污水的直接排放對溪流生態造成多樣性的影響，包含水溫上升、植物光合作用的降低、水棲物種死亡等，並進一步改變溪流之生態系。如今水體的模式建模已成為水資源管理的有效工具，合理的模擬可以輔助監測資料之不足，作為評判水域環境之依據。然而對於生態的模擬，釐清物種間的交互作用十分重要，結合穩定同位素分析或許是可行的方法。 故本研究以臺北陽明山的磺港溪進行生態與水質調查，分析溪流上、中、下游河段，受到不同程度之硫酸鹽溫泉排水影響，其對溪流生態系之改變。研究採集溪流河床生物膜、顆粒有機物、動物和植物樣本，分析其樣品之δ13C 及δ15N，並透過MIXSIAR (Bayesian Mixing Models in R)及SIBER(Stable Isotope Bayesian Ellipses in R)的分析方法，建立完整食物網，具體探討硫酸鹽污水對於溪流食物網結構之影響，並輸入AQUATOX模式進行模擬。 同位素研究結果顯示，在溪流受到硫酸鹽污染與pH改變之下，溪流中的物種食物來源隨河段越趨下游而越漸縮小，且各物種的食物貢獻率改變，食物網發生由下至上的影響，隨著污染程度加劇，多數消費者的同位素棲位(Isotopic Niche)變窄。 AQUATOX模式檢定與驗證的結果顯示，在多河段模擬中硝酸鹽和氨氮效率係數皆大於0.7，表示可以很合理的模擬水中的傳輸。藻類模擬中雖然下游稍有高估，其結果仍可以被接受。生物之模擬由於多河段模擬的限制，導致同位素建立的食物網無法輸入每個河段，使模擬結果欠佳。但在改成分段模擬後，使模擬的偏差降低，除中游魚類外，其餘物種實測與模擬之資料分布平均值與變異數均無顯著差異。證明將同位素建立的食物網輸入，能夠提升生態模型的模擬效果。 這些研究結果量化酸性廢水排放對水生生態造成的影響，並提供模擬方法，可作為酸性廢水的管制或改善措施的參考。

The direct discharge of hot spring wastewater has various impacts on stream ecosystems, including rising water temperatures, reduced plant photosynthesis, and the death of aquatic species, which further alter the stream ecosystem. Nowadays, water body modeling has become an effective tool for water resource management. Reasonable simulations can supplement insufficient monitoring data and serve as a basis for evaluating water environments. However, for ecological simulations, it is crucial to clarify the interactions between species. Combining stable isotope analysis may be a feasible method. This study conducts an ecological and water quality investigation on Huanggang Stream in Yangmingshan, Taipei. It analyzes the upstream, midstream, and downstream sections of the stream, which are affected by varying degrees of sulfate hot spring discharge and its impact on the stream ecosystem. The study collects samples of streambed biofilm, particulate organic matter, animals, and plants, and analyzes their δ13C and δ15N values. Using MIXSIAR (Bayesian Mixing Models in R) and SIBER (Stable Isotope Bayesian Ellipses in R) analysis methods, a complete food web is established to specifically explore the impact of sulfate wastewater on the stream food web structure. The data is then input into the AQUATOX model. The results of the study indicate that under sulfate pollution and pH changes, the food sources for species in the stream progressively diminish from upstream to downstream sections. Additionally, the food contribution rates for various species change, leading to a bottom-up effect on the food web. As the pollution level intensifies, the isotopic niche of most consumers narrows. This highlights how increased sulfate pollution constrains the diversity of available food sources and alters the ecological interactions within the stream ecosystem. The AQUATOX model calibration and validation results indicate that the efficiency coefficients for nitrate and ammonia nitrogen exceed 0.7 in multi-reach simulations, indicating a reasonable simulation of in-stream transport. While algal simulations slightly overestimated downstream values, the results remain acceptable. Due to the limitations of multi-reach simulations, the isotopically constructed food web could not be input for each reach, resulting in suboptimal outcomes. However, segment-based simulations reduced biases, showing no significant differences in the mean and variance of observed and simulated data for most species except midstream fish. This demonstrates that incorporating isotopically constructed food webs can effectively enhance the ecological model's simulation accuracy. These research results quantify the impact of acidic wastewater discharge on aquatic ecosystems and provide a simulation method that can serve as a reference for the regulation or improvement of acidic wastewater.