Temporary flood water storage in agricultural areas in the Middle Tisza river basin - Hungary

The Tisza river (966 km long, draining 157,218 km2 across about 14.4 million people, the Danube's longest tributary, average discharge 794 m3/s) flows from the Carpathians through Hungary's Great Pannonian Plain. Floods originate from fast mountain runoff accumulating in lowlands, worsened by deforestation, soil sealing and changing precipitation. 19th-century river regulation shortened the river by over 400 km and cut floodplain area by more than 90%, with dikes raised progressively as peak flood levels rose (753 cm in 1876, 909 cm in 1970, 1,040 cm in 2000). Flood dikes today total 2,850 km, protecting 16,000 km2 of the Tisza's 47,000 km2 Hungarian catchment; further climate-driven peak-level rises are projected, and pure dike reinforcement was judged prohibitively expensive (a 1999 World Bank-funded study estimated remaining upgrade costs at HUF 175 billion, about 700 million EUR at 1999 rates). Four severe floods in 1998-2001, including a 2001 dike rupture, exceeded historical peak levels, prompting a 4-year study (VITUKI, 2006) that found increased uncertainty and higher expected future flood levels.

Adopt a cost-effective flood-protection strategy for the middle Tisza basin able to cope with changing hydrological conditions and rising peak flows, tailored to local climatic variability and hydrological characteristics.

Following the 1998-2000 floods, government plans in 2000 and 2004 reinforced dikes (740 km, then a further 550 km planned) before a broader 2004 Act introduced reactivating former floodplain territory and building temporary flood reservoirs ('polders') alongside dike strengthening and restoring flood-channel run-off capacity, targeting up to 721 million m3 of long-term reservoir capacity. Agricultural land is used as controlled temporary flood storage during extreme events, supplemented by farmer compensation (an upfront lump sum for inconvenience/value loss, plus event-based damage compensation). The first polder opened in 2009; the remaining five were completed 2010-2015 with national and EU funding; one polder was used successfully during a 2010 flood. Hydrological modelling (Ungvari and Kis, 2022) shows using multiple polders simultaneously for major floods further reduces risk versus a single polder, and that additional polders remain cost-effective.

The overall solution cost around 260 million EUR, implemented with European Regional Development Fund and Cohesion Fund contributions. An ex-post cost-benefit analysis (Koncsos, 2006) found the 6-reservoir scenario, without further dike modification, substantially reduces risk versus no intervention at relatively low initial cost, and that further flood-defence investment is economically justified. A 2022 updated analysis (Ungvari and Kis) found using most reservoirs economically justified even for 20-30-year floods, i.e. more frequent use than the original 100-year design basis, raising the question of whether current agricultural land use in reservoir areas should shift toward forest to accommodate more frequent flooding use.

The strategy has proven effective, scalable and flexible for flood-risk reduction with considerable downstream benefit, but farmers and landowners were not involved in designing the strategy or its operating rules, limiting stakeholder acceptance. Compensation-scheme problems include inadequate coverage of soil-rehabilitation and production-cycle-disruption costs (especially for high-value crops), compensation processing taking up to a year, and high year-to-year unpredictability of compensation costs for the national budget -- issues likely to intensify as flood frequency rises.