I noticed a quite interesting paradox. In modeling the impact of temperature on mortality, DLNMs (Distributed Lag Non-linear Models) are considered a kind of gold standard. I understand the allure, as they neatly model the delayed impact of low temperatures increasing the spread of respiratory infections or of heatwaves, when an initial mortality spike is followed by subsequent below-trendline mortality.
There are a few known problems, like DLNMs’ tendency to correctly match spikes of mortality during heatwaves; thus, when extrapolating, they achieve spectacularly high numbers that later completely overshoot observational data.
Nevertheless, I notice an even more serious problem: they assume the relationship remains effectively constant or, in fancier models, assume some modest adaptation. By crunching larger datasets on seasonality, I noticed consistent shifts in mortality seasonality completely overshadowing any impact of climate change. For assumptions to hold, seasonality should have remained mostly constant over decades until starting to shift in the 1970s due to a changing warming trend. Yet, historical data (STL-extracted cycles) show the opposite. Any ideas? (Especially how to recalibrate DLNMs to match historical data?)
Article Anthropocene mortality cycle convergence: Global pathogen sp...
some researches "incorporate climate change effects into stochastic mortality models" that is "combine a stochastic mortality model with a climate epidemiology model"
Preprint Impacts of Climate Change on Mortality: An extrapolation of ...
https://arxiv.org/abs/2406.02054
however, most of the available / considered datasets are coming only from the northern hemisphere seasonality ...
Phil Geis "Are these not coincidental?" They are much too regular for random fluctuations. In prior, single-location works one could wonder what unique process was going on, while in my work I combined a dozen of countries and there is consistent pattern how those cycles have been evolving in over a century. On top of that, I could decompose the cycles on 3 PCA dimensions, 2 of them having quite clear link with climate (cold winters; hot summers; and another being something like changes of timing of lagging influenza deaths mixed up with apparent faze shift related to local timing of coldest and hottest month). However, there is a plot twist, that while for each period there is noticeable linkage between climate and cycles, the relationship changes over decades. (this findings somewhat challenges most climate change impact projections, which operate under assumption that those relations are effectively constant)
"On what basis do you assume these to be cause and effect?" Climate -> different pattern of seasonality of mortality? (I really don't think that it may be the other way around)
Steftcho P. Dokov Trying to incorporate stochastic link, could be a partial workaround to have the true trajectory covered. The problem for me is that what we observe does not appear to be anything like random walk, but rather quite consistent trends like adaptations to changing weather pattern (for which in modeling so far best fit appear to be quite fast, like matter of a few years) or general, long term tendency to subtle decrease in vulnerability to both high and low temperature.
- Badges
- Science topic