31 October 2018 in Journal of Engineering and Technological Sciences
DWDestika Agustina Widiawan
Destika Agustina Widiawan
+2
This paper discusses the result of the development of a hydro-meteorological hazard early warning system (H-MHEWS) that combines weather prediction from Weather Research and Forecasting (WRF) and the hydrometeorological hazard index from the National Disaster Management Authority (BNPB), Indonesia. In its current development phase, the hazards that H-MHEWS predicts are floods, landslides, and extreme weather events. Potential hazard indices are obtained by using an overlay approach and resampling so that the data have a 100-m spatial resolution. All indices are classified into 4 status categories: "No alert", "Advisory", "Watch", and "Warning". Flood potential is produced by overlaying rainfall prediction at 3-hour intervals with the flood index. Landslide potential is produced by overlaying rainfall prediction with the landslide index. Extreme weather potential is divided into 3 categories, i.e. heavy rain, strong winds, and extreme ocean waves. The whole prediction is dynamic, following weather predictions at 3-hour intervals. The hazard prediction results will trigger a 'Warning' alert in case of emergency status. This alert will be set up in a notification system to make it easier for the user to identify the most dangerous hydrometeorological hazard events.
1 November 2019 in IOP Conference Series: Earth and Environmental Science
In this study, model of deforestation is expressed as a function of the direct causes, each of these expressed as a function of the indirect causes. The socio-economic growth variables, such as population and GDP growth, were used as indirect causes of deforestation. The model assumed that deforestation was caused by wood consumption, forest product export, conversion to cropland, and forest fire. This study also calculated the elasticity of forest fire deforestation with respect to the population based on forest fire data between 1991 and 2000. Finally, it analysed the carbon emissions due to deforestation rate in Indonesian forest.
1 April 2020 in IOP Conference Series: Materials Science and Engineering
Hydrometeorological hazard early warning system has been developed in order to support public society in anticipating the extreme weather in Indonesia. Meanwhile, the early warning system in Indonesia still produced locally and need arduous human resources to process the enormous size data to have a good quality early warning. Therefore, in order to increase the effectivity, efficiency, and efficacy of the early warning system to produce the desired result, the automatic process was needed. This study will explain further the mechanism of process automation from the weather forecast to the output of warning displayed in an information system. The automation process combines several scripting in Python and batch script for GIS processing, as well as process scheduling using task scheduler. The output of extreme weather warning is in form of JSON files stored in organized folder directory. The automated process proven to be more useful in increasing the effectivity, efficiency, and efficacy of the early warning system to produce the desired result and later the automated process could be enhanced further to boost the lead time and reducing the arduous task in producing the early warning system.
1 June 2020 in IOP Conference Series: Earth and Environmental Science
In the past 5 years, the issue of forest and peatland fires in Indonesia has received a major concern from various parties. As one of the prevention efforts, the Government of Indonesia seeks to prevent the ignition of forest and peatland fires both anthropogenic and naturally as a result of prolonged drought. These efforts could be carried out well through the implementation of remote sensing observations and through predictive technology on the position of the hotspot. However, it should be noted that historical data that have been used in those predictive technology on these hotspots are not yet verified and validated with fire occurrences and burned area footprints. Therefore, in this study, verification of historical fires hotspot will be conducted with the occurrence of wildfires based on land use satellite imagery to validate the events. This verification test will use overlay and spatial statistical methods. based on the results of the verification test, a number of hotspots were found that were not in accordance with the incidence of forest fires especially in some area that neared agricultural area and factories. Thus, in the implementation of the use of hotspot data for predictions of forest fires, it is necessary to verify the suitability of the land and the incidence of forest fires.