Next-level vegetation health index forecasting: A ConvLSTM study using MODIS Time Series.

The Vegetation Health Index (VHI) is a metric used to assess the health and condition of vegetation, based on satellite-derived data. It offers a comprehensive indicator of stress or vigor, commonly used in agriculture, ecology, and environmental monitoring for forecasting changes in vegetation health. Despite its advantages, there are few studies on forecasting VHI as a future projection, particularly using up-to-date and effective machine learning methods. Hence, the primary objective of this study is to forecast VHI values by utilizing remotely sensed images. To achieve this objective, the study proposes employing a combined Convolutional Neural Network (CNN) and a specific type of Recurrent Neural Network (RNN) called Long Short-Term Memory (LSTM), known as ConvLSTM. The VHI time series images are calculated based on the Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST) data obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS) aboard the Terra and Aqua satellites. In addition to the traditional image-based calculation, the study suggests using global minimum and global maximum values (global scale) of NDVI and LST time series for calculating the VHI. The results of the study showed that the ConvLSTM with a 1-layer structure generally provided better forecasts than 2-layer and 3-layer structures. The average Root Mean Square Error (RMSE) values for the 1-step, 2-step, and 3-step ahead VHI forecasts were 0.025, 0.026, and 0.026, respectively, with each step representing an 8-day forecast horizon. Moreover, the proposed global scale model using the applied ConvLSTM structures outperformed the traditional VHI calculation method.

Prediction of MODIS land surface temperature using new hybrid models based on spatial interpolation techniques and deep learning models.

Land surface temperature (LST) prediction is of great importance for climate change, ecology, environmental and industrial studies. These studies require accurate LST map predictions considering both spatial and temporal dynamics. In this study, multilayer perceptron (MLP), long short-term memory (LSTM) and an integrated machine learning model, namely Convolutional LSTM (ConvLSTM), were utilized for one step ahead LST prediction. Data were gathered from 1-day (MYD11A1) and 8-day composite (MYD11A2) Moderate Resolution Imaging Spectroradiometer (MODIS) sensors, which have 1-km × 1-km spatial resolution. Considering the inability of MODIS sensors to provide LST data under cloudy conditions, Inverse DISTANCE WEIGHTING (IDW), natural neighbor (NN), and cubic spline (C) methods were used to overcome the missing pixel problem. The proposed methods were tested over the Northern part of Adana province, Turkey, and the performances of the models were quantitatively evaluated through performance measures, namely, Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The selected datasets range from 01 January 2017 to 01 November 2020 and from 01 January 2015 to 01 November 2020 for daily LST and 8-day composite LST, respectively. While 60% of the datasets were used as training set, the remaining 40% were used as validation (20%) and test (20%) sets. RMSE maps were generated to evaluate the pixelwise performance of the proposed method. On the other hand, the best average RMSE and MAE for the daily test set were obtained from the combination of ConvLSTM and NN (NN-ConvLSTM) as 3.62 °C and 2.85 °C, respectively, while they were acquired 3.57 °C and 2.69 oC from the combination of MLP and NN (NN-MLP) for the 8-day composite LST test set. The results revealed that the proposed hybrid models could be used for one step ahead spatiotemporal prediction of LST data.

Google Earth Engine based spatio-temporal analysis of air pollutants before and during the first wave COVID-19 outbreak over Turkey via remote sensing.

Air pollution is one of the vital problems for the sustainability of cities and public health. The lockdown caused by the COVID-19 outbreak has become a natural laboratory, enabling to investigate the impact of human/industrial activities on the air pollution. In this study, we investigated the spatio-temporal density of TROPOMI-based nitrogen dioxide (NO2) and sulfur dioxide (SO2) products, and MODIS-derived Aerosol Optical Depth (AOD) from January 2019 to September 2020 (also covering the first wave of the COVID-19) over Turkey using Google Earth Engine (GEE). The results showed a significant decrease in NO2 and AOD, while SO2 unchanged and had slightly higher concentrations in some regions during the lockdown compared to 2019. The relationship between air pollutants and meteorological parameters during the lockdown showed that air temperature and pressure were highly correlated with air pollutants, unlike precipitation and wind speed. Moreover, Purchasing Managers' Index (PMI) data, indicator of economic/industrial activities, also provided poor correlation with air pollutants. TROPOMI-based NO2 and SO2 were compared with station-based pollutants for three sites (suburban, urban, and urban-traffic classes) in Istanbul, revealing 0.83, 0.70 and 0.65 correlation coefficients for NO2, respectively, while SO2 showed no significant correlation. Besides, AOD data were validated using two AERONET sites providing 0.86 and 0.82 correlation coefficients. Overall, the satellite-based data provided significant outcomes for the spatio-temporal evaluation of air quality, especially during the first wave of the COVID-19 lockdown.

The acquisition of impervious surface area from LANDSAT 8 satellite sensor data using urban indices: a comparative analysis.

Rapid and irregular urbanization is an essential issue in terms of environmental assessment and management. The dynamics of landscape patterns should be observed and analyzed by local authorities for a sustainable environment. The aim of this study is to determine which spectral urban index, originated from old Landsat missions, represents impervious area better when new generation Earth observation satellite Landsat 8 data are used. Two datasets of Landsat 8, acquired on 2 September 2013 and 10 September 2016, were utilized to investigate the consistency of the results. In this study, commonly used urban indices namely normalized difference built-up index (NDBI), index-based built-up index (IBI), urban index (UI), and enhanced built-up and bareness index (EBBI) were utilized to extract impervious areas. The accuracy assessment of urban indices was conducted by comparing the results with pan-sharpened images, which were classified using maximum likelihood classification (MLC) method. The kappa values of MLC, IBI, NDBI, EBBI, and UI for 2013 dataset were 0.89, 0.79, 0.71, 0.59, and 0.49, respectively, and the kappa values of MLC, IBI, NDBI, EBBI, and UI for 2016 dataset were 0.90, 0.78, 0.70, 0.56, and 0.47, respectively. In addition, area information was extracted from indices and classified images, and the obtained outcomes showed that IBI presented better results than the other urban indices, and UI extracted impervious areas worse than the other indices in both selected cases. Consequently, Landsat 8 satellite data can be considered as an important source to extract and monitor impervious surfaces for the sustainable development of cities.

Evaluation of spatio-temporal variability in Land Surface Temperature: A case study of Zonguldak, Turkey.

The aim of this study is to analyze spatio-temporal variability in Land Surface Temperature (LST) in and around the city of Zonguldak as a result of the growing urbanization and industrialization during the last decade. Three Landsat 5 data and one Landsat 8 data acquired on different dates were exploited in acquiring LST maps utilizing mono-window algorithm. The outcomes obtained from this study indicate that there exists a significant temperature rise in the region for the time period between 1986 and 2015. Some cross sections were selected in order to examine the relationship between the land use and LST changes in more detail. The mean LST difference between 1986 and 2015 in ERDEMIR iron and steel plant (6.8 °C), forestland (3 °C), city and town centers (4.2 °C), municipal rubbish tip (-3.9 °C), coal dump site (12.2 °C), and power plants' region (7 °C) were presented. In addition, the results indicated that the mean LST difference between forestland and city centers was approximately 5 °C, and the difference between forestland and industrial enterprises was almost 8 °C for all years. Spatio-temporal variability in LST in Zonguldak was examined in that study and due to the increase in LST, policy makers and urban planners should consider LST and urban heat island parameters for sustainable development.