Related Project Review

Probabilistic Risk Modelling Review

CLIMADA is a software tool designed to model natural disasters and their impacts on human populations, infrastructure, and the environment[1]. It combines climate and weather data, demographic and infrastructure information, and geospatial data to simulate the impact of various natural disasters, such as floods, hurricanes, and wildfires. CLIMADA is used by decision-makers, planners, and researchers to assess the potential risk and impact of natural disasters on a particular region and to evaluate different mitigation and adaptation strategies[2]. Its goal is to provide information to help decision-makers better prepare for and respond to natural disasters. CLIMADA provides coverage for extreme-weather hazards, not low-impact ones.

Low-Impact High-Frequency Events

Our project functionality primarily lies within the impact prediction of low-impact hazardous events, as such while trying to find the state-of-the-art for it, we realised that there aren’t any ones in full-scale usage[3, 4]. Although CLIMADA is well documented, given the complexity of the model, it is very difficult to use as it does not contain video tutorials or a sophisticated user manual, especially when trying to develop new functions on it. Moreover, the exceedance calculation of CLIMADA relied on the frequency of events directly from data, which is not included in our data source. As such, we decided to create a simple, easy-to-use model to allow global coverage.

Technology Review

Programming Language - Python

We chose to use Python as our primary programming language because it offered us a range of powerful tools and resources for performing data analysis, manipulation, and visualization. Specifically, we relied on libraries such as NumPy and Pandas for data manipulation, BeautifulSoup for web-scraping in order to download datasets and Matplotlib for data visualization. In terms of skills, everyone on our team was most skilled using Python as such choosing it was a no-brainer.

Libraries - Pandas, Numpy and Beautiful Soup

Beautiful Soup was our library of choice in order to download all the datasets and parse XML Files into CSV as it is the most well-documented library for the purpose, which made it easier to use.

We chose to use Pandas and NumPy for a variety of reasons, such as the need for efficient and powerful data analysis and manipulation tools. These libraries can handle large datasets and provide numerous functions for data filtering, aggregation, grouping, merging, and more[5]. Additionally, the user-friendly API of Pandas and NumPy makes them easy to learn and use. Their integration with other popular Python libraries for data analysis and visualisation also makes them a convenient choice for data scientists and analysts. Moreover, the open-source nature of Pandas and NumPy, combined with their large and active community of developers and users, means that there are plenty of resources and support available online[6]. High Performance: Pandas and NumPy are both built on top of low-level, high-performance libraries like C and Fortran[7], which makes them very fast and efficient when working with large datasets.

Summary of technical decisions

Programming Language: Python

Library for Downloading and Converting: Beautiful Soup

Library for Data Analysis: Pandas and Numpy

LIbrary for Visualisation: Matplotlib

References

  1. Aznar-Siguan, G. and Bresch, D. N., 2019: CLIMADA v1: a global weather and climate risk assessment platform, Geosci. Model Dev., 12, 3085-3097, https://doi.org/10.5194/gmd-12-3085-2019
  2. Bresch, D. N. and Aznar-Siguan, G., 2021: CLIMADA v1.4.1: towards a globally consistent adaptation options appraisal tool, Geosci. Model Dev., 14, 351-363, https://doi.org/10.5194/gmd-14-351-2021
  3. Rana, A., Zhu, Q., Detken, A., Whalley, K., and Castet, C.: Strengthening climate-resilient development and transformation in Viet Nam, Climatic Change, 170, 4, https://doi.org/10.1007/s10584-021-03290-y, 2022.
  4. Consultores en Riesgos y Desastres, PROBABILISTIC MODELLING OF NATURAL RISKS AT THE GLOBAL LEVEL: THE HYBRID LOSS EXCEEDANCE CURVE, https://www.preventionweb.net/english/hyogo/gar/2011/en/bgdocs/ERN-AL_2011.pdf, 2011.
  5. Pandas. pandas documentation Package overview. [Online]. 2023 [Accessed 23 March 2023]. Available from: https://pandas.pydata.org/docs/getting_started/overview.html
  6. Numpy. NumPy: the absolute basics for beginners. [Online]. 2023 [Accessed 23 March 2023]. Available from: https://numpy.org/doc/stable/user/absolute_beginners.html
  7. J. D. Hunter, "Matplotlib: A 2D Graphics Environment", Computing in Science & Engineering, vol. 9, no. 3, pp. 90-95, 2007.