Spatiotemporal Analysis of Dissolved Organic Carbon and Nitrate in Waters of a Forested Catchment Using Wavelet Analysis

Abstract

Understanding natural controls on N and C biogeochemical cycles is important to estimate human impacts on these cycles. This study examined the spatiotemporal relationships between time series of weekly monitored stream and groundwater N and C (assessed by NO 3 − and dissolved organic C [DOC]) in the forested Wüstebach catchment (Germany). In addition to traditional correlation analysis, we applied wavelet transform coherence (WTC) analysis to study variations in the correlation and lag time between the N and C time series for different time scales. Median transit times were used to connect hydrologic and water chemistry data. We defined three stream-water groups: (i) subsurface runoff dominated locations with strong seasonal fluctuations in concentrations, short transit times, and strong negative C/N correlations with short time lags, (ii) groundwater dominated locations, with weaker seasonal fluctuations, longer transit times, and weaker C/N correlations with lags of several months, and (iii) intermediate locations, with moderate seasonal fluctuations, moderate transit times, and strong C/N correlations with short time lags. Water transit times could be identified as key drivers for the C/N relationship and we conclude that C and N transport in stream water can be explained by mixing of groundwater and subsurface runoff. Complemented by transit times and the hydrochemical time series, WTC analysis allowed us to discriminate between different water sources (groundwater vs. subsurface runoff). In conclusion, we found that in time series studies of hydrochemical data, e.g., DOC and NO 3 − , WTC analysis can be a viable tool to identify spatiotemporally dependent relationships in catchments.Abbreviations: COI, cone of influence; CWT, continuous wavelet transform; DOC, dissolved organic carbon; MedTT, median transit time; WTC, wavelet transform coherence; XWT, cross wavelet transform.The cycles of nutrients such as N and C are closely linked. For example, N saturation is linked to C limitation in the soil for microbial processes (Kopáček et al., 2013). In aqueous ecosystems, increasing NO 3 − concentrations (Galloway et al., 2008) and widespread significant increases in the concentrations of dissolved organic C (DOC) in streams during the last two decades have become a major global issue (e.g., Evans et al., 2005;Galloway et al., 2008;Vitousek et al., 1997). Newly created reactive N (biologically or photochemically active) contributes to environmental problems such as air pollution, eutrophication, soil acidification, and climate change (Vitousek et al., 1997;Galloway et al., 2008). The environmental and therefore also social consequences require detailed spatiotemporal models to predict the fate of reactive N across environments (e.g., Mulholland et al., 2008;Taylor and Townsend, 2010;Barnes and Raymond, 2010). For this purpose, it might be important to understand causal links to the spatiotemporal patterns of DOC. Core Ideas• WTC analysis was used to elucidate the non-stationary C/N relationship at different time...

Size, age, renewal, and discharge of groundwater carbon

“…Because of the diversity of landforms and geology covered, we believe these groundwater C values are likely substantially similar to the concentrations and distribution beneath other continents. Measurements of groundwater DOC (e.g., Huang et al 2015, Thayalakumaran et al 2015, Weigand et al 2017 and DIC (e.g., Chaillou et al 2014, Samanta et al 2015, Cao et al 2016) under other countries and continents, although rare, are in the same ranges as those in our survey. We determined the average C content at various depths by smoothing the depth versus groundwater age relationship using locally weighted sequential smoothing (Cleveland 1979) as described earlier.…”

Spectral Decomposition Reveals New Perspectives on CO2Concentration Patterns and Soil‐Stream Linkages

“…Conventional approaches to resolving aquatic transport problems are based on assessments of hydrochemical observations in the time domain. However, novel approaches transforming the problem into the frequency domain have emerged, allowing to analyze temporal patterns (Crawford et al, 2017;Kirchner & Neal, 2013;Schmidt et al, 2018;Weigand et al, 2017;Yan et al, 2017) and provide quantitative estimates of transport rates (Godsey et al, 2010;Guan et al, 2011;Riml & Wörman, 2015) as functions of the inherent time scales of the observed dynamics. The Fourier representation is a way to complement time domain analyses that unlocks new possibilities to utilize high-frequency hydrochemical observations.…”

Effects of Deforestation on Water Flow in the Vadose Zone

“…Carbon stored in soil and plant residues is receiving considerable interest, because they present a long term carbon sink that can be affected by climate and land use change [57]. This, in turn, can affect the transport of Dissolved Organic Carbon (DOC) through the vadose zone (e.g., [58]). Clearly, the transport of DOC is closely linked to its transit time as it reflects the potential of soils to buffer DOC [59,60].…”