Seasonal and spatial variations of both levels and characteristics of airborne bacteria and fungi in various types of non-complaint indoor environments and their outdoors were investigated. Bioaerosol samples were collected by the single-stage Andersen sampler during the winter and summer seasons in Ankara, Turkey. Indoor and outdoor temperature, relative humidity (RH) and CO₂ concentrations were also monitored online during the sampling. Bacteria levels in apartments were found to be considerably elevated when people were present. Significant relationships between bioaerosol levels and RH and CO₂ concentrations ( p<0.05) were found. Bacteria levels exhibited a seasonal variation, while fungi levels did not, probably due to lower fungi concentrations occurrence as a result of lower RH values (median: 30%). Measured low levels of fungi in this study may be due to geographical conditions, climatic factors, and other environmental conditions. The indoor to outdoor (I/O) ratio for bacteria levels were found to be significantly higher than 1.0, while indoor and outdoor fungi levels were similar. Winter to summer (W/S) ratios in the sampling site groups varied on a large scale for both culturable fungi concentrations (0.24–19.59) and total bacteria count (0.16–6.59). The most prominent bacteria were Micrococcus spp., Staphylococcus spp. and Bacillus spp., while the most predominant fungi were Penicillium spp., Aspergillus spp. and Cladosporium spp.

Turkey is located in the heart of complex transition geography between Eurasia and the Middle East. In the grand scheme, the so-called eastern Mediterranean Basin is located almost in the middle of the dusty belt, and is a hot spot of climate change. The downstream location of dust-carrying winds from close desert sources reveals Turkey as an open plane to particulate matter exposure throughout the year. In order to clarify this phenomenon, this paper aims to determine the desert dust climatology of Turkey via CALIPSO onboard Lidar. This prominent instrument enables us to understand clouds, aerosols and their types, and related climatic systems, with its valuable products. In this study, a 9-year CALIPSO-derived pure dust product dataset was formed to explain horizontal and vertical distributions, transport heights and case incidences. The results indicated that the pure dust extinction coefficient increased as the location shifted from west to east. Moreover, in the same direction of west to east, the dominant spring months changed to summer and autumn. Mountain range systems surrounding Anatolia were the main obstacles against lofted and buoyant dust particles travelling to northern latitudes. Even if high ridges accumulated mass load on the southern slopes, they also enabled elevated particles to reach the ground level of the inner cities.

This paper investigates the effect of the architectural design of deep learning models in combination with a feature engineering approach considering the temporal variation in the features in the case of tropospheric ozone forecasting. Although deep neural network models have shown successful results by extracting features automatically from raw data, their performance in the domain of air quality forecasting is influenced by different feature analysis approaches and model architectures. This paper proposes a simple but effective analysis of tropospheric ozone time series data that can reveal temporal phases of the ozone evolution process and assist neural network models to reflect these temporal variations. We demonstrate that addressing the ozone evolution phases when developing the model architecture improves the performance of deep neural network models. As a result, we evaluated our approach on the CNN model and showed that not only does it improve the performance of the CNN model, but also that the CNN model in combination with our approach boosts the performance of the other deep neural network models such as LSTM. The development of the CNN, LSTM-CNN, and CNN-LSTM models using the proposed approach improved the prediction performance of the models by 3.58%, 1.68%, and 3.37%, respectively.