The challenge of unprecedented floods and droughts in risk management

Risk management has reduced vulnerability to floods and droughts globally1,2, yet their impacts are still increasing3. An improved understanding of the causes of changing impacts is therefore needed, but has been hampered by a lack of empirical data4,5. On the basis of a global dataset of 45 pairs of events that occurred within the same area, we show that risk management generally reduces the impacts of floods and droughts but faces difficulties in reducing the impacts of unprecedented events of a magnitude not previously experienced. If the second event was much more hazardous than the first, its impact was almost always higher. This is because management was not designed to deal with such extreme events: for example, they exceeded the design levels of levees and reservoirs. In two success stories, the impact of the second, more hazardous, event was lower, as a result of improved risk management governance and high investment in integrated management. The observed difficulty of managing unprecedented events is alarming, given that more extreme hydrological events are projected owing to climate change3.

• Supplementary Table 2: Representative examples from flood and drought paired events of quantitative variables and textual descriptions corresponding to the five classes of change ranging from large decrease (-2) to large increase (+2) from the first event used as baseline to the second event of a pair.In case of quantitative comparisons, a change of less than 50% is treated as small, and above 50% as large.
• Supplementary Table 3: Indicators-of-change and sub-indicators indicate large change (-2/2), small change (-1/1) or no change (0) from the first event used as baseline to the second event of a pair.Coping capacity, which is the ability of communities using available skills and resources, to manage an event was insufficient due to a lack of funding (insurance, risk transfer), resources or skills Low or lacking public flood compensation to individuals and businesses

Management shortcomings Problems with water management infrastructure
Water management infrastructures such as levees, reservoirs, sewage systems, etc. failed or did not work optimally during an event due to deficits in maintenance, sub-optimal design, etc.

Number of levee breaches
Lack of water in reservoirs, insufficient storage capacity Non-structural risk management shortcomings Non-structural risk management measures, e.g.spatial planning that avoids increase of exposure in hazardprone areas and private property level risk mitigation measures were not optimally implemented Lack of hazard and risk maps Ineffective water use restrictions

References
This are the additional references of Supplementary Supplementary Table 2: Representative examples from flood and drought paired events of quantitative variables and textual descriptions corresponding to the five classes of change ranging from large decrease (-2) to large increase (+2) from the first event used as baseline to the second event of a pair.In case of quantitative comparisons, a change of less than 50% is treated as small, and above 50% as large.

Number of fatalities Direct economic impacts Indirect impacts Intangible impacts
Large decrease (-2)

Impact indicators for droughts Direct economic impacts Indirect impacts Intangible impacts
About 35,000 job losses in agriculture, estimated 50,000 people pushed below poverty line due to job losses and food price inflation, drop in tourism (Ziervogel 2019; City of Cape Town 2019; WWF 2018) (ID 44) NA Drivers

of impact Hazard indicators for floods Antecedent conditions Precipitation/weather severity Severity of flood
nd floodNo rainfall in the last previous 5 days.Some inlets clogged by leaves (BCASA, 2018) (ID 12) April-May accumulated precipitation between 240-300 mm (preliminary data, Agriculture and Agri-Food Canada, n.d.) (ID 41) 4.10 m surge (Adnan et al. 2019) plus high tide (UNDP, 2010) (ID 20) Small increase (+1) st flood Late winter conditions and snowpack were considered average for the basin for May-April.Heavy localized rainfall events happening at the same time as snowmelt led to high soil saturation and river flows in early April.However, the primary driver of flooding was rainfall runoff (McNeil, 2019; ORRPB, 2018).(ID 41) 62.5 mm (Areal average of 3-day precipitation maxima for German part of the Upper Danube catchment) (Schröter et al., 2015) (ID 15) 7,700 m³/s peak discharge at gauge Achleiten (~HQ50) (HND 2021); 1,081 cm water level at gauge Passau; 10,250 m³/s peak discharge at Korneuburg/Vienna (Blöschl et al., 2013) (ID 15)