Nitrate is one of the focal water quality indices in aquatic systems. However, proper estimation of nitrate concentration is a complicated task. In this article, capabilities of deep neural networks (DNN) and conventional artificial neural networks (ANN) to model and predict nitrate concentration in the Klokot River, Bosnia and Herzegovina were investigated. The measured data includes nitrate and pH values. Different scenarios were considered as the potential models for DNN and ANN structuring. The result showed DNN and ANN networks are incapable of precisely modeling the nitrate concentration in the Klokot River based on the designed scenarios. In addition, DNN was slightly superior to ANN in different terms of estimation accuracy.

• Based on the previous study of the heat-treated wood at 0-3 mm surface layer, this study focuses on the transition of thermal modification intensity on 160-220 °C heat-treated poplar from surface to core layers. The color change was evaluated by CIELAB, and surface hardness was detected via Shore D (HD) and pressing ball method (H R ); furthermore, the FT-IR was applied to detect the thermal degradation of wood components. The results show that the degradation of cell wall components in the surface layer of heat-treated poplar wood is greater than that in the core layers

• Wood heat treatment is an environmentally friendly method, and the heat-treated wood properties are closely related to thermal modification intensity. This study focuses on the 0-3 mm surface layer (SL) of poplar wood heat treated at 160~220 °C. The modification intensity, including surface color, hardness, chemical component and morphological changes of the SL, was evaluated. The findings of this research showed that the color difference of the poplar wood before and after heat treatment (ΔE *1 ) increased; the color difference between up-surface and down-surface of the SL (ΔE *2 ) also increased with the treatment temperature. Consequently, the surface hardness (H R ) decreased with the increase of treatment intensity. When the treatment temperature was higher than 160 °C, the up-surface and down-surface of the SL were statistically different in color. Chemical component analysis revealed that the heat treatment degrades wood components, especially the hemicellulose, and correlation analysis showed a significant correlation between the change rate of hemicellulose and the ΔE * 1 or H R value; the prediction functions have been established at a high confidence level of 0.99. Overall, the thermal modification intensity of the heat-treated surface layer (SL) of poplar wood varies, and the H R and ΔE *1 value could be used to characterize and predict the modification intensity and degree of thermal degradation of the surface layer of heat-treated poplar wood.