The Weather File Is Out of Date

To understand the problem, understand what a weather file is and where it comes from. When an engineer simulates a building's energy use or checks whether it overheats, the simulation needs to know the weather the building experiences - so it is fed a weather file, most commonly a typical meteorological year (TMY): a synthetic year assembled from real historical measurements to represent 'typical' conditions at that location - hourly temperature, humidity, solar radiation, wind, over a representative year. These files are the foundation of building performance analysis and energy codes worldwide. Crucially, they are built by looking backward: a TMY is stitched together from perhaps twenty or thirty years of *past* recorded weather. Embedded in that construction is a deep assumption - stationarity - that the climate is stable enough that a synthesis of the recent past is a fair description of the near future. Under a stable climate, that assumption held well enough for decades, and building analysis quietly relied on it.

That assumption has now broken. The climate is warming rapidly and non-linearly: average temperatures are rising, heatwaves are becoming more frequent, more intense and longer, humidity and rainfall patterns are shifting, and extremes that were once rare are becoming common. A weather file assembled from the last few decades therefore describes a climate that is already in the past - it systematically understates the heat, and the extremes, that even today's buildings face, and it diverges further from reality every year. A cooling system sized to a TMY may be too small for the heatwaves that now actually arrive; a naturally-ventilated building judged 'comfortable' by the old data may now spend dangerous hours overheating; an energy prediction anchored to historical weather may be badly wrong. This is not a subtle academic point - it is a live, worsening mismatch between the data we design to and the world buildings actually inhabit. The first move of climate analytics is simply to see this clearly: the standard baseline is stale, the past is no longer a safe guide, and continuing to design to yesterday's typical weather is a choice with real, and increasingly dangerous, consequences.

The stale-baseline problem would matter even for a short-lived object; for a building it is acute, because of a simple mismatch of timescales. Buildings are extraordinarily long-lived. A house, an office, a school, a hospital built today is expected to serve for fifty, eighty, often more than a hundred years - and across that whole life it must keep the people inside comfortable and, in a warming world increasingly, *safe*. Now set that lifespan against the pace of climate change. The climate a building will experience is not a fixed condition but a moving target that shifts through every decade of its life: a building finished in the near future will spend its middle age in the climate of the mid-century and its old age in the climate of the century's end, each markedly hotter and more extreme than today, and vastly hotter than the historical average in its weather file. So there is a double gap: the weather file is already behind *today's* climate, and today's climate is itself far milder than the climate the building will face in its later decades. Designing to the file means designing for the coolest, gentlest conditions the building will ever see - at the very start of a life that only gets hotter.

This reframes what 'good design' means. A building optimised precisely for its historical weather file is optimised for a world that ends almost as soon as it opens. A building that merely meets today's code, checked against today's stale data, may be code-compliant and still fail its occupants in the heatwaves of the decades to come. The right question is not 'is this comfortable and efficient in the typical weather of the recent past?' but 'will this remain comfortable, safe and efficient across the range of climates it will actually experience over its long life, including the hotter and more extreme ones?' That is a fundamentally different design question, and answering it requires looking not backward at historical weather but forward at *future* weather - which is exactly what climate analytics sets out to do, and exactly where the honest difficulty begins, because the future, unlike the past, cannot be measured, only projected under uncertainty.

Climate analytics answers the stale-baseline problem by bringing the *future* into design - and an honest course is clear from the first page that this introduces a deep, irreducible uncertainty that must be handled with discipline, not bluffed past. The method is genuinely powerful: climate scientists produce projections of future climate from physical models; these are used to create future weather files (often by 'morphing' a historical file to reflect projected changes) for a future decade like the 2050s or 2080s; and building performance simulation then tests a design against that future weather - revealing whether it will overheat, how its energy use will shift, whether it stays survivable in a future heatwave. This lets a designer see a building's future performance while it can still be changed, which is a real and important capability. But the honesty must be equally clear: a future weather file is a scenario, not a forecast. How hot it actually gets depends on how much greenhouse gas humanity emits - which is a matter of human choices no one can predict - so projections are made under different emissions scenarios that diverge enormously by late century. On top of that, climate models disagree with one another, and translating global projections to a specific site adds further uncertainty. The result is not a single confident future but a *range* of plausible futures, sometimes a wide one.

This creates a specific, serious danger the course will return to: false precision. A future weather file arrives as a neat, hour-by-hour data set that looks exactly as precise as a historical measurement - but it is nothing of the sort; it is one plausible scenario carrying deep uncertainty, and treating its numbers as a prediction ('the building will be 2.3 degrees warmer in 2050') is a serious error that can mislead as badly as ignoring the future entirely. The competent stance is neither the denialist's (design to the historical file and ignore the future) nor the false-precision technocrat's (simulate a 2050 file and treat the output as fact), but the disciplined one: use future projections to understand the *direction, range and severity* of the risk, design for resilience and robustness across a range of plausible futures rather than optimising for one predicted number, and always remember you are managing risk under uncertainty, not predicting the weather of a distant decade. And one more honesty runs underneath: adaptation is not enough. Designing buildings to cope with a hotter world is essential, but you cannot adapt your way out of unlimited warming - beyond some level of heating no building keeps people safe - so climate-resilient design must sit alongside, never replace, cutting emissions in the first place.

This course builds climate-analytics and future-weather literacy as a practical, honest design skill. You will start with designing for a changing climate - the stale weather file, climate vs weather, the landscape, the limits (Module 0); then why future weather matters - a building outlives its climate, the overheating problem, resilience and survivability, the caveats (Module 1); climate and weather data - weather files and TMY, sources of climate data, reading the data, data quality and uncertainty (Module 2); climate projections and future weather - how projections work, emissions scenarios, morphing future weather files, the uncertainty of the future (Module 3); climate analysis for design - climate analysis basics, degree-days and comfort, sun/wind/humidity, the climate study (Module 4); simulating performance - building performance simulation, overheating analysis, energy under future weather, extreme-event analysis (Module 5); designing for resilience - passive survivability, adaptive and robust design, cooling in a hotter world, flooding/wind/extremes (Module 6); beyond the single building - urban heat and microclimate, the vulnerable and equity, adaptation vs mitigation, resilience at scale (Module 7); making it real - the analysis workflow, tools and data, integrating into design, working with specialists (Module 8); reality, limits and honesty - climate-washing, false precision, when the model misleads, adaptation is not enough (Module 9); and practice and the future - the designer's role, getting started, India, becoming climate-literate (Module 10).

One firm boundary runs through all of it. Climate analytics and future-weather simulation are tools to understand risk and inform design under uncertainty - they produce scenarios, not forecasts, and estimates, not guarantees. This course teaches the principles and design judgement, and defers every binding result - the building-physics, energy, thermal-comfort, structural and climate-risk engineering of any design, and any compliance or life-safety determination - to qualified building-physics, energy, structural and climate-risk engineers, verified data and validated tools, and the governing codes and standards (the National Building Code of India, the Energy Conservation Building Code, the relevant IS standards and recognised methods). Any weather file, projection or result named here is illustrative, scenario-dependent and uncertain. Studio Matrx is free and not-for-profit, and this course is written to be rigorous and honest - not a resilience sales pitch but a clear, critical grounding in designing for the climate a building will face, mindful of the Indian context where heat is already deadly and worsening, where cooling demand is surging, where humidity, monsoon and flooding compound the risk, where a vast and vulnerable population is most exposed and least able to afford protection, and where designing survivable buildings is a matter of real moral urgency. Understand the stale baseline, the long life of a building against a fast-changing climate, the genuine power of future-weather analysis and its deep uncertainty, and above all the disciplines of designing for a range not a number and pairing adaptation with mitigation - and you will be able to design buildings that keep people safe in the world that is actually coming.