This thesis estimates the economic effects of temperature, precipitation, and relative sea level rise (SLR) on regional economic growth in Europe from 1900 to 2015. A central starting point is the model developed by Burke, Hsiang, et al. (2015) (BHM), which provided the first global evidence that aggregate economic growth responds non‑linearly to temperature, with growth peaking at an annual average temperature of 13 °C and declining sharply at higher temperatures. Rising temperatures can reduce productivity and agricultural yields, especially in warmer regions (Somanathan et al., 2021). Shifts in precipitation patterns can cause longer dry periods or heavier rainfall, increasing the risk of floods and stressing water systems and infrastructure (Kotz et al., 2022; Malhi et al., 2021). SLR increases the risk of flooding and loss of land in low‑lying coastal areas (Cortés Arbués et al., 2024; Chatzivasileiadis et al., 2023).

This thesis addresses key gaps in the climate econometric literature by re‑analysing and extending the BHM model with a regional dataset and additional climate variables for over a century. In doing so, it moves beyond theory‑based models by relying on observed historical data to estimate the economic effects of climate variables with greater spatial and temporal detail. This research uses a quantitative approach, applying an econometric model to regional (NUTS‑2 level) climate data from the Climatic Research Unit, University of East Anglia (n.d.) and economic data by Rosés et al. (2021). Following Clemens (2017), it performs a re‑analysis and extension of the BHM model to examine whether its findings remain valid, testing the stability of the concave relationship between temperature and economic growth while also including precipitation and SLR to capture the combined effects of multiple climate variables.

The results from this research support the overall concave shape of the relation between temperature and economic growth, meaning that even over a longer time frame, at the regional level, and with the inclusion of an additional climate variable, temperature and economic growth are linked in a non‑linear way, with growth peaking at an optimal temperature of 11.8 °C. Temperature has the strongest effect on economic growth among the climate variables included in the model. Precipitation does not show a statistically significant effect on its own, but it is jointly significant when included with other climate variables. SLR has a smaller (compared to temperature) but statistically significant impact, suggesting that it plays a meaningful role alongside temperature in shaping economic outcomes. However, the estimated coefficients for temperature in this research are approximately six times larger than those in the model by BHM, indicating stronger temperature sensitivity at the regional level. The sensitivity analysis further shows that these coefficients are not stable, particularly across different benchmark years, suggesting that the estimated turning point is shaped by historical context and should not be interpreted as a fixed economic threshold.

Future research should build on this work by using micro‑level climate and economic data to examine the effects of seasonal variation, heatwaves, extreme rainfall, and other short‑term events (Kotz et al., 2024; Somanathan et al., 2021). This could be combined with models that include adaptation processes or delayed responses to climate change (Mérel et al., 2021). The use of only twelve benchmark years in this study limits the ability to identify such short‑run effects. Increasing temporal resolution would improve understanding of both short‑term shocks and long‑term structural patterns.

It will also be essential to better understand how climate variables interact. Temperature is a known driver of both precipitation and SLR, through its influence on atmospheric moisture and ice melt (Malhi et al., 2021). These dependencies should be explicitly modelled in future studies to avoid misattributing indirect effects or underestimating compound risks (Kotz et al., 2022). Finally, moving beyond the reduced‑form approach by BHM would allow for more flexible models that capture causal mechanisms and account for the possibility of multiple optima, thresholds, or plateaus that a fixed quadratic form may overlook. These steps are essential to ensure that econometric models can support meaningful and context‑specific design of climate policies.

The Mekong Delta in Vietnam, one of the world’s most fertile and ecologically rich deltas, faces environmental challenges that threaten its biodiversity and local communities. Historically, the delta’s extensive mangrove forests provided natural coastal protection, but in recent decades, agricultural expansion, aquaculture, and infrastructure developments have degraded these ecosystems.
This research focuses on Bac Liêu, a region acutely affected by these changes. With diminishing mangrove buffers, local vulnerabilities to environmental hazards have increased, putting pressure on sea defenses. In response, the Dutch government and Vietnamese partners have introduced the “Mekong Living Lab,” an initiative for in-field research that promotes mangrove restoration and sustainable coastal management.
Conducted by TU Delft students, this study contributes to the Living Lab’s goals by exploring the causes of mangrove decline in Bac Liêu. Combining interviews with local residents and field data on coastal profiles, this multidisciplinary approach seeks to safeguard the ecological and economic future of the Mekong Delta.
The study suggests an integrated approach within the Living Lab framework, emphasizing research, showcasing, and education to bridge hydraulic, ecological, and socio-economic perspectives. Priority recommendations include continuous cross-sectional measurements, sediment retention analysis, stakeholder engagement strategies, and further interdisciplinary studies on mangrove viability. These initiatives aim to align technical insights with stakeholder needs, advancing observation-driven solutions for Bac Liêu’s mangrove ecosystems.