Michael Scheuerer

This report describes the research related to use case 3 within pilot ♯6 of the FAME project: Embedding Climatic Predictions in Property Insurance Products. The changing climate poses risks to the global economy in various ways. In addition to the immediate impacts on the real estate sector, e.g., due to changing living conditions, and impacts on the insurance sector as a result of more frequent and intense weather-related disasters, it calls for the transition to a greener and more sustainable economy, which comes with a significant price tag. While the latter risk cannot be directly related to specific weather events, an increasing body of research suggests that climate risk indices constructed through textual analysis of newspaper articles are able to represent the different types of risks and can thus help build climate risk hedge portfolios. In this report, we document our own efforts to construct such a news-based climate risk index and analyze whether climate-focused funds perform differently than a benchmark representing the general market. Our risk index is constructed from an analysis of climate change-related newspaper articles in the New York Times which were further filtered by utilizing large language model (LLM) with zero-shot classification. Both full and subcategory-based indices are defined at a daily resolution and were aggregated to both weekly and monthly resolution in order to analyze impacts on stock market returns at different time scales. Our analysis, however, does not show a clear and robust link between the climate risk index and the stock market returns of climate-focused funds compared to the general market. A more refined selection of ’green’ vs. ’brown’ stocks may be required to see significant, climate risk index-dependent differences in performance that can be exploited for the purpose of portfolio optimization.

A linear regression model is developed to link anomalies of streamflow to anomalies of precipitation amounts and temperature with the goal of making multi-decadal streamflow projections based on CMIP6 multi-model simulations. Regression coefficients estimated separately for each catchment and each month show physically implausible spatial patterns and indicate issues with overfitting. An alternative approach is therefore explored in which all regression coefficients are estimated simultaneously through a neural network that retains the original linear model structure, but uses embeddings to map each combination of catchment and month to a set of regression coefficients. The model is demonstrated over a set of catchments in Brazil, where the estimated relationships are used to make streamflow projections for the next decades based on CMIP6 multi-model simulations. It yields physically more plausible relationships between streamflow, precipitation amounts, and temperature for our study area than the locally fitted regression models. The resulting projections indicate reduced streamflow over northern, north-eastern, central, and south-eastern Brazil, especially for the austral spring and summer season. The signal is less clear during austral winter. In southern Brazil, an increase in streamflow is expected.

Improving probabilistic streamflow forecasts is critical for a multitude of water-oriented applications. Errors in water forecasts arise from several sources, one of which is the driving meteorology. Meteorological forecasts are often statistically postprocessed before being input into hydrologic models. Shifts toward ensemble weather prediction systems have propelled advances in ensemble postprocessing, providing an opportunity to enhance probabilistic water forecasting. This study’s purpose is to implement and evaluate the impact of coupling state-of-the-art precipitation ensemble postprocessing techniques with the process-based, spatially distributed National Water Model (NWM). The postprocessing has two steps: First, precipitation is calibrated using a censored, shifted, gamma distribution approach, and second, it is reordered using an ensemble copula coupling technique. The NWM focuses on flood forecasting but to date has only been run with time-lagged ensemble weather forecasts. We implement the NWM in a medium-range (∼7 day) ensemble forecasting mode for several rain-dominated catchments in Northern California during an extremely wet water year, when advanced warning of heavy precipitation and streamflow could have been useful. Postprocessing enhances NWM streamflow forecasts in terms of ensemble spread and accuracy, improving underestimation. Precipitation (streamflow) was generally skillful out to day 4 (7), including heavy precipitation (>75 mm) and relatively high-flow thresholds, but less consistently for the most extreme streamflow. These results suggest that NWM ensembles could be warranted for priority basins with relatively predictable weather phenomena, though there are trade-offs with hydrological model complexity and ensemble forecasting. This study can inform the NOAA-led Next Generation Water Resources Modeling Framework, which will need to consider how to integrate meteorological postprocessing and ensemble techniques. Significance Statement The purpose of this study is to implement and evaluate the impact of statistically corrected ensemble weather predictions, with a focus on precipitation, on ensemble streamflow forecasts. Our results for a very wet year in California show improved performance in terms of both precipitation and streamflow, in particular reducing underestimation. This is important because advanced warning of potential heavy precipitation and streamflow is critical to improve society’s readiness for adverse weather and water impacts.

This report describes the research related to use case 2 within pilot ♯6 of the FAME project: Embedding Climatic Predictions in Property Insurance Products. In a changing climate, it is expected that the frequency and/or intensity of nat ural disasters will change. This has impacts on several sectors of the economy, perhaps the most obvious being disaster-related damages and losses affecting the profitability of insurance companies. To explore and quantify this link, we analyze the impact of major natural disasters on the volatility of insurance stocks. Our working hypothesis is that the value at risk (VaR) of these stocks increases in the wake of a disaster as a result of an expected spike in claims, loss in value of assets held by insurers, etc. To test this hypothesis, we analyze data on the economic losses associated with major disasters in Europe since January 2000, and we build a statistical model to model their effect on the volatility (and thereby VaR) of the different stock prices. For storms, one of the two most damaging types of disasters in Europe, we also calculate projections of a loss index based on climate model simulations of daily maximum wind speeds, and we analyze how the regionally aggregated loss index is projected to change in the future. Combined with the statistical model that links storm losses to the VaR of insurance stocks, these loss index projections could give us an idea about climate-related changes in VaR. Our analysis, however, does not show a clear and unequivocal link between nat ural disasters and insurance stock volatility. While some major disasters are followed by a drop in the price of certain insurance stocks, the signal is inconsistent and not statistically significant. This is in line with other studies in the literature, which find a significant link only for some regions and some types of disasters, but it leads us to conclude that a robust projection of climate-related changes in VaR cannot be obtained with the data at hand.

This report describes the research related to use case 1 within pilot ♯6 of the FAME project: Embedding Climatic Predictions in Property Insurance Products. For a dataset with house value statistics over a high-resolution grid over California (USA), a hedonic regression model was fitted that explains the median house value over a grid cell through factors like population density, median income, and ocean proximity. Additional variability can be explained by a component in the regression model that quantifies the reduction in house value due to frequent episodes with extreme heat at this grid cell. Daily mean temperature simulations from several regional climate models are then statistically further downscaled to the resolution of the house price grid, and the projected increase in the number of days per year with excessive heat over the next decades is studied. When combining this information with the fitted regression model, the associated projected decrease of house values can be calculated.

This works proposes a probabilistic framework for rainy season onset forecasts over Greater Horn of Africa derived from bias-corrected, long range, multi-model ensemble precipitation forecasts. A careful analysis of the contribution of the diferent forecast systems to the overall multi-model skill shows that the improvement over the best performing individual model can largely be explained by the increased ensemble size. An alternative way of increasing ensemble size by blending a single model ensemble with climatology is explored and demonstrated to yield better probabilistic forecasts than the multi-model ensemble. Both reliability and skill of the probabilistic forecasts are better for OND onset than for MAM and JJAS onset where forecasts are found to be late biased and have only minimal skill relative to climatology. The insights gained in this study will help enhance operational subseasonal-to-seasonal forecasting in the GHA region.

Postprocessing is a critical step in attaining calibrated and reliable probabilistic forecast output from numerical weather prediction models. A novel deep learning framework is proposed to postprocess 20 years of 7- and 14-day precipitation accumulation reforecasts from the Global Ensemble Forecast System at subseasonal time scales (week 1, week 2, and combined weeks 3–4 forecasts) over the contiguous United States. The network builds upon previous studies and is a combination of three parallel-trained components suitable for subseasonal prediction. The first is a ResUnet architecture which learns nonlinear relationships between binned observed precipitation and input images of weather and geographical variables. The second conditions the network to the month-of-year via a feature-wise linear modulation (FiLM) layer. The third helps the network learn when to revert the forecast to that of climatology. The RUFCO (named for its components ResUnet, FiLM, and Climatological-Offramp) forecasts are compared against raw and climatological forecasts as well as those from a state-of-the-art distributional regression postprocessing model, “censored, shifted gamma distribution (CSGD),” and a simple bias-corrected model. At week 1, every method exhibited a competitive advantage over climatological forecasts. At week 2, RUFCO generated forecasts with statistically significant improvement over climatology at 82%–94% of the domain, beating CSGD’s coverage of 76%–90% of the grid points. At week 3, RUFCO’s skillful coverage was 65%–85%, while CSGDs dropped to only 12%–37%. At the longer lead times, RUFCO achieved the highest domain-averaged skill scores across seasons. However, the network tends to “smooth” forecast skill, making it less competitive with CSGD in limited areas with strongly spatially varying biases. Significance Statement Precipitation accumulation forecasts 1, 2, and 3–4 weeks in advance are increasingly in-demand for a variety of decision-making applications around hydrologic forecasting, flood and drought awareness, and wildfire preparedness. However, raw forecasts from numerical weather prediction systems have errors that hinder skill. Postprocessing methods remove those errors and provide more reliable and skillful forecasts. We show that a new neural network technique is an effective and competitive postprocessing tool compared to more traditional techniques.

In the postprocessing of ensemble forecasts of weather variables, it is standard practice to first calibrate the forecasts in a univariate setting, before reconstructing multivariate ensembles that have a correct covariability in space, time, and across variables, via so-called “reordering” methods. Within this framework though, postprocessors cannot fully extract the skill of the raw forecast that may exist at larger scales. A multi-temporal-scale modulation mechanism for precipitation is here presented, which aims at improving the forecasts over different accumulation periods, and which can be coupled with any univariate calibration and multivariate reordering techniques. The idea, originally known under the term “canonical events,” has been implemented for more than a decade in the Meteorological Ensemble Forecast Processor (MEFP), a component of the U.S. National Weather Service’s (NWS) Hydrologic Ensemble Forecast Service (HEFS), although users were left with material in the gray literature. This paper proposes a formal description of the mechanism and studies its intrinsic connection with the multivariate reordering process. The verification of modulated and unmodulated forecasts, when coupled with two popular methods for reordering, the Schaake shuffle and ensemble copula coupling (ECC), is performed on 11 Californian basins, on both precipitation and streamflow. Results demonstrate the clear benefit of the multi-temporal-scale modulation, in particular on multiday total streamflow. However, the relative gain depends on the method used for reordering, with more benefits expected when this latter method is not able to reconstruct an adequate temporal structure on the calibrated precipitation forecasts.

Tropical cyclones are extreme events with enormous and devastating consequences to life, property, and our economies. As a result, large-scale efforts have been devoted to improving tropical cyclone forecasts with lead times ranging from a few days to months. More recently, subseasonal forecasts (e.g., 2–6-week lead time) of tropical cyclones have received greater attention. Here, we study whether bias-corrected, subseasonal tropical cyclone reforecasts of the GEFS and ECMWF models are skillful in the Atlantic basin. We focus on the peak hurricane season, July–November, using the reforecast years 2000–19. Model reforecasts of accumulated cyclone energy (ACE) are produced, and validated, for lead times of 1–2 and 3–4 weeks. Week-1–2 forecasts are substantially more skillful than a 31-day moving-window climatology, while week-3–4 forecasts still exhibit positive skill throughout much of the hurricane season. Furthermore, the skill of the combination of the two models is found to be an improvement with respect to either individual model. In addition to the GEFS and ECMWF model reforecasts, we develop a statistical modeling framework that solely relies on daily sea surface temperatures. The reforecasts of ACE from this statistical model are capable of producing better skill than the GEFS or ECMWF model, individually, and it can be leveraged to further enhance the model combination reforecast skill for the 3–4-week lead time.

Agricultural food production and natural ecological systems depend on a range of seasonal climate indicators that describe seasonal patterns in climatological conditions. This article proposes a probabilistic forecasting framework for predicting the end of the freeze-free season, or the time to a mean daily near-surface air temperature below 0°C (referred to here as hard freeze). The forecasting framework is based on the multimodel seasonal forecast ensemble provided by the Copernicus Climate Data Store and uses techniques from survival analysis for time-to-event data. The original mean daily temperature forecasts are statistically postprocessed and downscaled with a mean and variance correction of each model system before the time-to-event forecast is constructed. In a case study for a region in Fennoscandia covering Norway for the period 1993–2020, the proposed forecasts are found to outperform a climatology forecast from an observation-based data product at locations where the average predicted time to hard freeze is less than 40 days after the initialization date of the forecast on October 1.

Wildfire guidance two weeks ahead is needed for strategic planning of fire mitigation and suppression. However, fire forecasts driven by meteorological forecasts from numerical weather prediction models inherently suffer from systematic biases. This study uses several statistical-postprocessing methods to correct these biases and increase the skill of ensemble fire forecasts over the contiguous United States 8–14 days ahead. We train and validate the postprocessing models on 20 years of European Centre for Medium-Range Weather Forecasts (ECMWF) reforecasts and ERA5 reanalysis data for 11 meteorological variables related to fire, such as surface temperature, wind speed, relative humidity, cloud cover, and precipitation. The calibrated variables are then input to the Global ECMWF Fire Forecast (GEFF) system to produce probabilistic forecasts of daily fire indicators, which characterize the relationships between fuels, weather, and topography. Skill scores show that the postprocessed forecasts overall have greater positive skill at days 8–14 relative to raw and climatological forecasts. It is shown that the postprocessed forecasts are more reliable at predicting above- and below-normal probabilities of various fire indicators than the raw forecasts and that the greatest skill for days 8–14 is achieved by aggregating forecast days together.