Role of aerosols in deciding pH of rainwater

Nature of precipitation – alkaline or acidic – depends upon the concentration of major water soluble inorganic gaseous and soil derived particulates dissolved in it. If concentration of cations is higher than that of anions, the precipitation becomes alkaline and vise-a-versa. pH is the main parameter indicating the nature of precipitation. If pH of the solution is < 5.65, it is acidic and > 5.65, alkaline, in the pH scale ranging between 0 & 14. Difference in the chemical composition of rainwater having pH in the alkaline range and acidic range has been studied in this paper. For this purpose, precipitation chemistry data of Allahabad, Jodhpur, Mohanbari and Nagpur for the period 1988-97 have been considered. APWM & S.D. have been worked out. Precipitation chemistry data considering APWM values in acidic range and in alkaline range have been separated and compared. Coefficients of correlation have been calculated in possible cation-anion combinations. It is seen that the lowest pH values (monthly mean) have been recorded during 1997 – MHB (3.77), NGP (4.12), both in acidic range. % Frequency of occurrences of pH values in acidic range was the highest at Mohanbari (66%) in the study period. Jodhpur recorded all pH values in alkaline range indicating influence of soil derived alkaline particulates. Among cations Ca+² has shown its dominance over all cations. Jodhpur has recorded the highest APWM value of Ca+² (9.27mg/lit) in 1997. Data have also been compared with WMO Laboratory simulated acid rain sample analysis data and other non-departmental data. Results are discussed in the paper.

Qmin: A machine learning-based application for mineral chemistry data processing and analysis

Mineral chemistry analysis is a valuable tool in several phases of mineralogy and mineral prospecting studies. This type of analysis can point out relevant information, such as concentration of the chemical element of interest in the analyzed phase and, thus, the predisposition of an area for a given commodity. Due to this, considerable amount of data has been generated, especially with the use of electron probe micro-analyzers (EPMA), either in research for academic purposes or in a typical prospecting campaign in the mineral industry. We have identified an efficiency gap when manually processing and analyzing mineral chemistry data, and thus, we envisage this research niche could benefit from the versatility brought by machine learning algorithms. In this paper, we present Qmin, an application that assists in increasing the efficiency of mineral chemistry data processing and analysis stages through automated routines. Our code benefits from a hierarchical structure of classifiers and regressors trained by a Random Forest algorithm developed on a filtered training database extracted from the GEOROC (Geochemistry of Rocks of the Oceans and Continents) repository, maintained by the Max Planck Institute for Chemistry. To test the robustness of our application, we applied a blind test with more than 11,000 mineral chemistry analyses compiled for diamond prospecting within the scope of the Diamante Brasil Project of the Geological Survey of Brazil. The blind test yielded a balanced classifier accuracy of ca. 99% for the minerals known by Qmin. Therefore, we highlight the potential of machine learning techniques in assisting the processing and analysis of mineral chemistry data.

Investigating the effect of lab bias on long-term stream chemistry data

Long-term (LT) stream chemistry studies are used to examine changes in and responses to the environment. Much of the data collected over long periods of time goes through changes in instrumentation, methods, and personnel potentially resulting in changing values. A data user must understand these measures of data quality through quality control (QC) results to know with certainty if trends are real or attributable to other factors. We used the Web of Science database search engine to search for LT stream chemistry studies. For each study, we then determined: record or study length; if QC was reported; and if QC was used. We found that 33% of papers reported QC in the method, and 12% presented the QC in the results. Next, we conducted a case study on 46 years of stream chemistry data to evaluate the data with and without the application of QC protocols from two watersheds (WS) at Coweeta Hydrologic Laboratory; WS 7; clear-cut in 1967–77 and adjacent WS 2 which serves as a reference. We focused on nitrogen and sulfur due to their importance in understanding the forest ecosystem response to disturbance (NO3) and acid deposition (SO4). We determined average annual dissolved inorganic nitrogen (DIN) export (NH4 + NO3 = DIN) using three methods for censoring values below the method detection limit (mdl): (1) the found value, (2) the value of zero, and (3) one-half the mdl value. We found that DIN export for WS 2/WS 7 was (1) 66.9/831.4 (g ha−1 yr−1), (2) 45.4/808.0 (g ha−1 yr−1), and (3) 72.1/823.2 (g ha−1 yr−1) using the three censoring methods, and that the export estimate was significantly different for WS 2 but not for WS 7 (P = 0.001). We found that on average stream NH4 concentrations were below the mdl 58% of the time until an instrument change in 1994 resulted in improved mdls resulting in fewer data points below detection. We found increased bias for stream SO4 concentration following an instrumentation change from segmented flow analysis to ion chromatography. As a result, stream SO4 concentration data that were bias-corrected declined more rapidly in WS 2 compared with non-bias-corrected data, but not in WS 7. We conclude that including QC results with LT data is essential to verify data validity and give the data user a full understanding of the results.

Beyond Data Validation – Advanced Strategies for Assessing Data Quality for Oil Spill Investigations

ABSTRACT Field sampling investigations in response to oil spill incidents are growing increasingly more complex with analytical data collected by a variety of interested parties over many years and with different investigative purposes. For the Deepwater Horizon (DWH) Oil Spill, the analytical chemistry data and toxicity study data were required to be validated in accordance with U.S. Environmental Protection Agency's (EPA's) data validation for Superfund program methods. The process of validating data according to EPA guidelines is a manual and time-consuming process focused on chemistry results for individual samples within a single data package to assess if data meet quality control criteria. In hindsight, the burden of validating all of the chemistry data appears to be excessive, and for some parameters unnecessary, which was costly and slowed the process of disseminating data. Depending on the data use (e.g., assessing human and ecological risk, qualitative oil tracking, or forensic fingerprinting), data validation may not be needed in every circumstance or for every data type. Publicly available water column, sediment, and oil chemistry analytical data associated with the DWH Oil Spill, obtained from the Gulf of Mexico Research Initiative Information and Data Cooperative data portal were evaluated to understand the impact, effort, accuracy, and benefit of the data validation process. Questions explored include: What data changed based on data validation reviews?How would these changes affect the associated data evaluation findings?Did data validation introduce additional errors?What data quality issues did the data validation process miss?What statistical and data analytical approaches would more efficiently identify potential data quality issues? Based on our evaluation of the chemical data associated with the DWH Oil Spill, new strategies to assess the quality of data associated with oil spill investigations will be presented.

Estimation of groundwater CO2 concentrations on a catchment scale using Random Forest

&lt;p&gt;Water-rock-interactions in the saturated and unsaturated zone govern the natural variability of CO&lt;sub&gt;2&lt;/sub&gt; in groundwater. However, anthropogenic pollutions such as excessive input of organic and inorganic fertilizers or sewage leakage can cause shifts in the carbonate-pH system in an aquifer. Additional dissolution of minerals and associated mobilization of harmful heavy metals are possible consequences. Anthropogenic groundwater pollution is especially an issue where a protective confining layer is absent. On the other hand, addressing an environmental hazard such as fertilizer input to a single parameter remain intricate due to the high number of possible competing reactions such as microbial-controlled redox reactions. To overcome these obstacles, machine learning based statistical methods become increasingly important.&lt;/p&gt;&lt;p&gt;This study attempt to predict the CO&lt;sub&gt;2 &lt;/sub&gt;concentration in groundwater from a multi-feature selection by using Random Forest. For this purpose, groundwater chemistry data (in situ measured bulk parameter, major ions, nutrients, trace elements and more) from more than 23000 wells and springs in Germany were collected and homogenized in a single database. Measured or calculated CO&lt;sub&gt;2 &lt;/sub&gt;concentrations&lt;sub&gt;&lt;/sub&gt;are used to train the Random Forest algorithm and later to validate model results. Beside chemistry data, features about hydrogeology, soil characteristics, land use land cover and climate factors serve as predictors to build the &amp;#8220;forest&amp;#8221;. The intention of this study is to establish comprehensive CO&lt;sub&gt;2 &lt;/sub&gt;predictions based on surface and climate features and to identify trends in local CO&lt;sub&gt;2 &lt;/sub&gt;production. Gained knowledge can be used as input for groundwater quality management processes and adaptation policies.&lt;/p&gt;