Abstracting past studies and synthesizing their spatial data in GIS for utilization in a study of flood risk in Attica region, Greece

From the perspective of a team of engineers that studies an area in order to estimate flood risk, a highly studied area presents both benefits and challenges. On the one hand, stands the benefit of accessibility to data and existing knowledge, while on the other the challenge of optimal analysis and utilization of such material.

The region of Attica, Greece, fits exactly the description of a highly studied area. It hosts the capital of the country, Athens, and close to half of the national population. For this reason, flood risk has been studied throughout the 20^th and 21^st century in various spatial scales, and using different methods and tools. In a contemporary study of flood risk with modern digital computational tools, we undertook the tasks of compiling, analyzing and mapping important information from those past studies.

In this work, we present the relevant challenges faced during the project “Earthquake, fire and flood risk assessment in the Region of Attica” as well as inferences for future studies of similar type. Following the previous step of our comprehensive methodology for the estimation and mapping of flood risk in Attica Region, we received as an input a set of existing studies for our study area. In the process of abstracting the spatial data from those studies, we grouped such data in three categories: spatial information in text, in CAD files and in GIS files. Each type of information presented its own type of challenges. Information in text was time-consuming to process, both because it necessitated reading the whole studies to be pinpointed, but also due to the requirement for manual extraction and conversion to GIS format. Spatial information found in CAD and GIS files presented mostly software-related challenges in managing to achieve a common representation scale and geospatial reference of the plans. Lastly, a challenge that was common for all types of data was the identification of whether the hydraulic works or terrain conditions that were presented in each study were actually present as such today. This required cross-methodological communication including both with the previous methodological steps of contact with the institutions that provided the studies, and with the next methodological steps of site visits for the confirmation of the recorded information.

Overall, we identify the following as crucial priorities for efficient abstracting and synthesizing the spatial data on flood risk in a highly studied area. Firstly, clarity in defining the types of data being searched for, secondly, the utilization of CAD and GIS software with interoperability functions and easy scaling and georeference functionalities, thirdly, the direct communication with the institutions that created or utilize the studies, and lastly, the realization of site visits to verify the recorded information.