Date set for the manuscript: Exploring the temporal dynamics of methane ebullition in a subtropical freshwater reservoir

<p>The dataset supports the findings of the manuscript entitled &lsquo;Exploring the Temporal Dynamics of Methane Ebullition in a Subtropical Freshwater Reservoir&rsquo;. The results of the manuscript are based on continuous in-situ measurements conducted at Passa&uacute;na Reservoir, located in the southern part of Brazil (South America). The monitoring was carried out from 2017 to 2020, during which a comprehensive set of environmental variables was obtained from various studies. The primary objective of the manuscript was to comprehend the temporal dynamics of ebullition flux in the reservoir. Therefore, time series data of ebullition and relevant environmental variables were analyzed across different time scales, ranging from minutes to daily resolutions. Statistical and data driven models were tested to predict ebullition at varying time scales, considering the influence of environmental variables. The findings are discussed in the manuscript.</p> <p>Here, xlsx files and Matlab scripts are provided. The xlsx files contain time series data for environmental variables (&#39;Environmental_Variables_TimeSeries&#39;) and ebullition flux (&#39;Ebullition_Flux_TimeSeries&#39;). Separate sheets were utilized for different time intervals, namely 5 minutes (dt=5min), 10 minutes (dt=10min), 1 hour (dt=1hr), and 1 day (dt=1d). Explanations and units are provided within the column labels, while &#39;NaN&#39; denotes missing data.</p> <p>Matlab scripts for all the empirical models that were tested, as outlined in the manuscript&#39;s supporting information (Tables S1 and S2), have been made available (detailed in the following table). These scripts were developed using MatLab R2023a.</p> <table> <tbody> <tr> <td> <p>File (.m)</p> </td> <td> <p>Description</p> </td> </tr> <tr> <td> <p><strong>Main scripts</strong></p> </td> </tr> <tr> <td> <p>TableS1_Empirical_Models_Literature</p> </td> <td> <p>Empirical models from the literature tested. (Table S1 in the manuscript)</p> </td> </tr> <tr> <td> <p>TableS2_FirstPart_Reffited_Models</p> </td> <td> <p>Empirical models from the literature refitted. (Table S2 in the manuscript)</p> </td> </tr> <tr> <td> <p>TableS2_SecondPart_New_Models</p> </td> <td> <p>New empirical models implemented. (Table S2 in the manuscript)</p> </td> </tr> <tr> <td> <p>TableS2_SecondPart_New_Models_GAM_Timescales</p> </td> <td> <p>Generalized additive models (GAM) applied to predict ebullition at different timescales. (Table S2 in the manuscript).</p> </td> </tr> <tr> <td> <p>PredictedR2_GAM_models</p> </td> <td> <p>Calculate the predicted R-squared for GAM.</p> </td> </tr> <tr> <td> <p><strong>Functions (need for the main scripts)</strong></p> </td> </tr> <tr> <td> <p>load_TimeSeries</p> </td> <td> <p>Import time series of environmental variables from excel sheets into MatLab.</p> </td> </tr> <tr> <td> <p>load_EbullitionTS</p> </td> <td> <p>Import time series of ebullition from excel sheets into MatLab.</p> </td> </tr> <tr> <td> <p>units_description</p> </td> <td> <p>Description containing units of the variables.</p> </td> </tr> <tr> <td> <p>Bin_xdata</p> </td> <td> <p>Creates data binning of X and Y based on X data.</p> </td> </tr> <tr> <td> <p>RelativeError</p> </td> <td> <p>Calculate the relative error of accumulated flux between measured and simulated.</p> </td> </tr> <tr> <td> <p>ANN5Neurons_Retrained_Deshmukh2014</p> </td> <td> <p>Trained Artificial Neural Network based on the ANN proposed by Deshmukh et al. (2014).</p> </td> </tr> <tr> <td> <p>ANN_Passauna_20Neurons</p> </td> <td> <p>Trained Artificial Neural Network with the addition of more environmental variables.</p> </td> </tr> </tbody> </table> <p>&nbsp;</p> <p><strong>Financial support</strong></p> <p>The field measurements were financed by the German Federal Ministry of Education and Research (BMBF, Grant 02WGR1431A), in the framework of the research project MuDak-WRM (https://www.mudak-wrm.kit.edu). This study was financed in part by the Coordena&ccedil;&atilde;o de Aperfei&ccedil;oamento de Pessoal de N&iacute;vel Superior &ndash; Brasil (CAPES) &ndash; Finance Code 001. Tobias Bleninger received productivity stipend from the National Council for Scientific and Technological Development (CNPq, grant no. 312211/2020-1, call no. 09/2020). Michael M&auml;nnich received productivity stipend from the National Council for Scientific and Technological Development (CNPq, grant no. 308744/2021-7, call no. 04/2021). Andreas Lorke received financial support from the German Research Foundation (DFG, grant number LO1150/16-1).</p>