Applying Machine Learning for Threshold Selection in Drought Early Warning System
نویسندگان
چکیده
This study investigates the relationship between Normalized Difference Vegetation Index (NDVI) and meteorological drought category to identify NDVI thresholds that correspond varying categories. The gridded evaluation was performed across a 34-year period from 1982 2016 on monthly time scale for Grassland Temperate regions in Australia. To label each grid inside climate zone, we use Australian Gridded Climate Dataset (AGCD) 120-year 1900 2020 calculate percentiles corresponding classification model takes data as input outputs of Then, propose threshold selection algorithm distinguish indicate boundary two adjacent performance is evaluated using accuracy metric, visual interpretation heat map. provides concept evaluate severity, well severity. results this demonstrate potential application toward early warning systems.
منابع مشابه
Evaluating the Application of a Financial Early Warning System in the Iranian Banking System
One of the significant problems of banks and investors in Iran is the lack of precise awareness about the financial performance of each bank and the roadmap for improving the conditions. Besides, the undesirable status of the financial performance of banks becomes evident only when the improvement of conditions is complicated. In this paper, a data mining-based early warning system (EWS) model ...
متن کاملDesign and Development of Early Warning System for Desertification and Land Degradation
Early warning systems are key components of strategies to reduce risk. This research, by adopting a systematic approach in the management of the risk of desertification and by including previously developed models and systems, offers an integrated efficient structure in terms of early warning for the risk of desertification as a pilot system for semi-arid areas of west Golestan Province in IRAN...
متن کاملIntegrated System of Early Warning, Monitoring and Drought Risk Analysis for Romania
Drought is one of the major natural calamities causing important damages to natural vegetation, agriculture and the society. The climate changes scenarios forecast the increase and intensification of extreme phenomena all over the globe. Romania is one of the most vulnerable countries, and has 7 million hectares located within areas affected by the risk of desertification. An integrated system ...
متن کاملData Mining Early Warning System Based on Virtual Learning Environment
In this paper, multi-angle analysis of quantitative data mining warning system is completed, by using software engineering method. The study on the core issue of early warning is conducted, during divided the learning process in VLE platforms into three stages: former learning, study of middle and after learning. The studying route of core issues is: under the mode of Windows Server Apache, MYS...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Climate
سال: 2022
ISSN: ['2225-1154']
DOI: https://doi.org/10.3390/cli10070097