Impact of long-lasting flood water on agricultural productivity: a case study of the May 2023 Emilia Romagna floods

Heavy rainfall events occurred in the Emilia-Romagna region in Northern Italy as a result of two major storms on May 2^nd and 17^th that led to the overflow of 22 rivers and triggered over 250 landslides. This event claimed 15 lives, forced 10 thousand people to evacuate and caused over 400 road closures. Due to a prior long-lasting winter drought and poor land use management that hampered effective water drainage, floodwaters stagnated for over a month in some areas, exacerbating the crisis. Over 40% of regional agricultural land was flooded leading to irreversible crop damage, in some instances, entire harvest loss. The objective of this study is to build a consistent and replicable methodology to quantify the agricultural damages and economic loss resulting from stagnated floodwater over cropland using the Emilia Romagna floods as a case study. The study emphasises the use of remote sensing data as a tool to achieve accurate impact estimates. Sentinel-1 SAR imagery is used to derive 10-meter resolution flood extent and duration maps at a revisit time of 3 to 6 days. The maps are matched with crop data available for the region from the iColt database and damages are computed as a function of ponded water duration and crop type as well as resistance to oxygen deprivation. The data, comprised of crop type, growing season and sowing date, allow for the characterization of the growth state of each crop at the time of flooding, implicitly providing insights on the probability of plant survival. The use of satellite-derived vegetation indices as markers for post-disaster crop recovery, with a focus on identifying crop-specific recovery rates and patterns is highlighted. This study highlights the need for collaborative efforts with key regional entities and can provide factual-hazard-based agricultural loss estimates to local institutions. These findings can guide targeted adaptation strategies, improve the spatial accuracy of loss assessment, and improve our comprehension of the aftermath of prolonged floods on agricultural output.