Assessing and predicting landfill surface temperature using remote sensing and an artificial neural network

Assessing and monitoring of the landfill temperature is of great importance, so as to assess environmental impacts of the landfill and to prevent fires that may lead to failure of the landfill operations. In this study, satellite images were used in mapping the landfill surface temperature (LFST) of Al Akeeder landfill site in Northern Jordan. Artificial Neural Network (ANN) model was developed to simulate and predict the LFST. Fifty-four Landsat satellite images on different dates covering the period 2000-2016 were collected and utilized after being subjected to correction of the thermal band. Multi-temporal thematic maps were developed for the landfill site and the LFST trends, patterns and magnitude were evaluated. Correlating LFST with the amount of the landfilled solid waste has resulted in a good correlation coefficient (r = 0.884), which implies that LFST can serve as an indicator of the amount of solid waste buried in the landfill. Solid waste amount, methane emitted as well as meteorological parameters, such as temperature, humidity, wind speed and evaporation, were used to simulate and predict the value of landfill LFST using ANN. Validation of the ANN has resulted in a good correlation between the predicted and calculated values of the LFST.