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

<p>Water-rock-interactions in the saturated and unsaturated zone govern the natural variability of CO<sub>2</sub> 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.</p><p>This study attempt to predict the CO<sub>2 </sub>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<sub>2 </sub>concentrations<sub></sub>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 “forest”. The intention of this study is to establish comprehensive CO<sub>2 </sub>predictions based on surface and climate features and to identify trends in local CO<sub>2 </sub>production. Gained knowledge can be used as input for groundwater quality management processes and adaptation policies.</p>

Impact of Urbanization on Sunshine Duration from 1987 to 2016 in Hangzhou City, China

Worldwide solar dimming from the 1960s to the 1980s has been widely recognized, but the occurrence of solar brightening since the late 1980s is still under debate—particularly in China. This study aims to properly examine the biases of urbanization in the observed sunshine duration series from 1987 to 2016 and explore the related driving factors based on five meteorological stations around Hangzhou City, China. The results inferred a weak and insignificant decreasing trend in annual mean sunshine duration (−0.09 h/d decade−1) from 1987 to 2016 in the Hangzhou region, indicating a solar dimming tendency. However, large differences in sunshine duration changes between rural, suburban, and urban stations were observed on the annual, seasonal, and monthly scales, which can be attributed to the varied urbanization effects. Using rural stations as a baseline, we found evident urbanization effects on the annual mean sunshine duration series at urban and suburban stations—particularly in the period of 2002–2016. The effects of urbanization on the annual mean sunshine duration trends during 1987−2016 were estimated to be −0.16 and −0.35 h/d decade−1 at suburban and urban stations, respectively. For urban stations, the strongest urbanization effect was observed in summer (−0.46 h/d decade−1) on the seasonal scale and in June (−0.63 h/d decade−1) on the monthly scale. The notable negative impact of urbanization on local solar radiation changes was closely related to the changes in anthropogenic pollutions, which largely reduced the estimations of solar radiation trends in the Hangzhou region. This result highlights the necessity to carefully consider urbanization impacts when analyzing the trend in regional solar radiation and designing cities for sustainable development.