Continuous measurements at Schwingbachtal Studienlandschaft in the growing seasons 2013-2014

The recent development of in-situ monitoring devices, such as UV-spectrometers, makes the study of short-term stream chemistry variation relevant, especially the study of diurnal cycles, which are not yet fully understood. Our study is based on high-frequency data from an agricultural catchment (Studienlandschaft Schwingbachtal, Germany). We propose a novel approach, i.e. the combination of cluster analysis and Linear Discriminant Analysis, to mine from these data nitrate behavior patterns. As a result, we observe a seasonality of nitrate diurnal cycles, that differs from the most common cycle seasonality described in the literature, i.e. pre-dawn peaks in spring. Our cycles appear in summer and the maximum and minimum shift to a later time in late summer/autumn. This is observed both for water- and energy-limited years, thus potentially stressing the role of evapotranspiration. This concluding hypothesis on the role of evapotranspiration on nitrate stream concentration, which was obtained through data mining, broadens the perspective on the diurnal cycling of stream nitrate concentrations.近年来，原位监测设备（如紫外分光光度计（UV-spectrometers））的发展，使得针对溪流短期化学变化的研究——尤其是尚未得到充分阐释的昼夜周期研究——具备了愈发重要的研究价值。本研究基于德国施温巴赫谷研究区（Studienlandschaft Schwingbachtal）这一农业集水区的高频监测数据，提出一种全新方法：将聚类分析（cluster analysis）与线性判别分析（Linear Discriminant Analysis）相结合，从上述数据中挖掘硝酸盐的行为模式。研究结果显示，硝酸盐昼夜周期存在季节特征，且与文献中记载的典型季节模式存在差异：现有文献普遍认为春季的硝酸盐昼夜峰值出现在黎明前，而本研究观测到的峰值多出现在夏季，且在夏末/秋末时节，峰值与谷值的出现时间均有所延后。这一现象在水分受限与能量受限年份中均被观测到，从而进一步凸显了蒸散发（evapotranspiration）的潜在作用。本研究通过数据挖掘得到的关于蒸散发对溪流硝酸盐浓度影响的总结性假说，为溪流硝酸盐浓度的昼夜周期研究拓宽了学术视角。