亞熱帶山地集水區河川新陳代謝之流域尺度模擬

Abstract

河川新陳代謝（stream metabolism）通常與淨生態系統生產力（net ecosystem productivity）息息相關並且為碳循環在生態圈中重要的途徑之一，被視為一能夠有效度量生態系統功能的方式，可以以河川中溶氧（dissolved oxygen）的日夜變化來表示。河川新陳代謝整合了許多不同的生態系統過程，例如它控制了河川中的光合作用與呼吸作用。而河川新陳代謝主要會受控於光合有效輻射（photosynthetic active radiation）、地貌、水文特徵、生物群系、營養鹽濃度和溫度等環境參數。然而在過去研究中鮮少著重於小山地集水區，目前我們對於臺灣河川新陳代謝的量測是嚴重不足的，也缺乏集水區尺度的河段推估，導致我們在河川連續體（river continuum）的概念中是不了解的。本研究以在北勢溪進行的河川新陳代謝觀測為基礎，透過整合上述的重要環境參數並將其涵蓋於空間河道網絡模型(Spatial Stream Network Models)中，以達到小山地集水區新陳代謝的模擬。結果顯示各站每月的日平均河川新陳代謝值擁有顯著的季節變動並介於-16.0 g O2 m-2 d-1至11.6 g O2 m-2 d-1之間，其中全站最低月平均值出現於一月（-16.0 g O2 m-2 d-1）；並在七月時達到頂峰（11.6 g O2 m-2 d-1），而水溫、pH值與不同的土地利用型態與河川新陳代謝呈現顯著相關性。此外，本研究展示了空間河道網絡模型於臺灣小山地集水區的可行性，並提供較非空間模型更優良的模擬表現，上游河道擁有較低的河川新陳代謝數值，顯示其河道可能屬於異營主導的，然而下游河道則逐漸轉變為自營主導的狀態。藉由本研究可增加對於臺灣小山地集水區河川新陳代謝運行機制的理解。

Stream metabolism, closely linked to net ecosystem productivity (NEP), is one of important routes of carbon cycle and aquatic ecosystem functions. Stream metabolism can be described as the diurnal variation of dissolved O2 (DO) concentration, in which photosynthetic active radiation (PAR), landscape setting, hydrologic regime, biome, nutrient concentrations and temperature regulate. However, few studies focused on stream metabolism in subtropical small mountainous rivers (SMRs), where the change of stream metabolism along river is unclear. Our study used the hourly DO records to estimate the local stream metabolism. Further the controlling factors supplementary with landscape metrics was introduced to Spatial Stream Network (SSN) models for simulating stream metabolism along rivers. Results showed that the monthly mean NEP in each study sites from 2018 to 2021 vary from -16.0 g O2 m-2 d-1 to 11.6 g O2 m-2 d-1 with strong seasonal variation. Specifically, the lowest monthly mean NEP was observed in January (-16.0 g O2 m-2 d-1) and reached the peak in July (11.6 g O2 m-2 d-1). Meanwhile, the water temperature, pH, and different land cover likely affect stream metabolism significantly. Additionally, this study demonstrates the applicability of SSN models, which outperformed NS models, for estimating stream metabolism in SMRs in Taiwan. The low and negative NEP indicates the headwater rivers are heterotrophic-dominated, but changes toward autotrophic-dominated downstream. The understanding of the mechanism of stream metabolism in SMRs, Taiwan could be increased through this study.