Typhoons strongly impact the structure and functioning of the forests, especially in the coastal areas in which typhoon-induced flooding imposes additional stress on the ecosystem via physical destruction and rising soil salinity. The impact of typhoons on forest ecosystems is becoming even more significant in the changing climate, which triggers atmospheric mechanisms that increase their frequency and intensity. This study investigates the resiliency of the Philippines’ forest areas (i.e., two selected forestry areas in Tacloban and Guiuan) in the aftermath of Super Typhoon Haiyan, which was followed by coastal flooding, as well as changes in ecosystem and biomass content using remote sensing. For this, we first evaluated the sensitivity of the normalized difference vegetation index (NDVI), green normalized difference vegetation index (GNDVI), and enhanced vegetation index (EVI) in detecting temporal changes in biomass content using very high-resolution satellite images. Then, employing three resilience concepts: amplitude, malleability, and elasticity, the most sensitive biomass index (i.e., NDVI) and digital elevation model (DEM) data were used to measure the resiliency of the Guiuan and Tacloban sites. We also applied a mean-variance analysis to extract and illustrate the shifts in the ecosystem status. The results show that despite a considerable biomass loss (57% in Guiuan and 46% in Tacloban), the Guiuan and Tacloban sites regained 80% and 70% of their initial biomass content within a year after the typhoon, respectively. However, the presence of canopy gaps in the Tacloban site makes it vulnerable to external stressors. Furthermore, the findings demonstrate that the study areas return to their initial states within two years. This indicates the high resiliency of those areas according to elasticity results. Moreover, the evaluation of typhoon impacts according to the elevation demonstrates that the elevation had a substantial impact on both damage severity and biomass recovery.

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.