Weather Architecture Jonathan Hill

Weather Architecture Jonathan Hill

I spent about three years working with this framework after a colleague pointed me at Jonathan Hill's research on climate-responsive building envelopes. It wasn't love at first sight. The early iterations were rough, the documentation scattered across academic papers and conference presentations rather than any single coherent manual. But once you get past the learning curve, it does something most other weather modeling approaches for architecture simply don't: it ties actual meteorological data directly to spatial design decisions rather than treating climate as an afterthought at the engineering stage. The core method is straightforward enough on paper. You take long-term weather data for a specific site, run it through Hill's layered analysis process, and map the results onto building geometry decisions. The layers cover solar exposure patterns, wind flow characteristics, thermal mass requirements, and natural ventilation potential. Each layer informs the next. Most people try to run all four simultaneously and end up with conflicting recommendations, so I usually recommend doing them sequentially and locking in each layer before moving forward.

Weather Architecture Jonathan Hill

Here is the practical workflow. Start with site selection and collect at least five years of local weather station data, preferably from the nearest official meteorological station. The closer the station to your actual build site, the more useful the data. If you are working in a location with no nearby station, you can interpolate from regional climate models, but the accuracy drops significantly and the later layers become unreliable. I learned that the hard way on a project in rural New Zealand where the nearest station was forty kilometers away and the microclimate was completely different due to proximity to coastal fog patterns. Layer one is the solar analysis. Hill's approach uses sun path diagrams overlaid with hourly insolation data to identify critical exposure windows. Mark which facades receive dominant solar gain during heating seasons versus cooling seasons. This isn't theoretical, I pulled standard sun path charts for a site in Toronto and immediately caught that a glazing specification I had been planning for the north facade would actually be receiving significant afternoon solar gain in late winter due to the terrain reflecting light off a nearby ridge. The standard model assumed flat terrain. Terrain correction is something the original papers mention in passing but almost nobody applies. Layer two covers wind dynamics. This is where the method gets interesting and also where most people mess up. You map prevailing wind directions and velocities by season, then cross-reference with your building's likely orientation and massing. The goal is identifying natural ventilation pathways and pressure zones around the structure. I used to skip the seasonal wind shift analysis and just look at annual averages. That works fine for temperate climates with consistent patterns. It failed completely on a coastal project where the summer and winter wind regimes were nearly opposite due to sea breeze cycles. Building the ventilation stack effect based on annual averages meant the system worked well six months out of the year and was actively counterproductive the other six.

Layer three is thermal mass placement. Once you know your solar exposure patterns and wind profiles, you determine where thermal mass should go and how much. Heavy materials on the solar-received sides for passive heating, lighter construction on the protected sides. The trick is balancing this against structural requirements and cost. Thermal mass isn't free, and over-specifying it can actually hurt performance if the material doesn't have time to release stored heat between cycles. I calculated thermal mass for a project and overshot by about thirty percent because I didn't account for the interior partition walls also contributing to thermal mass. The room temperatures stayed stable but the system had less capacity to absorb peak loads than I had designed for. Under-specification is equally dangerous but easier to avoid because you tend to overshoot when you are unsure. Layer four ties everything together with natural ventilation strategy. This layer depends on all the previous ones being correct. If your solar analysis was wrong, your ventilation strategy will be wrong. If your wind mapping missed a seasonal pattern, your stack effect calculations will be off. The interdependency is the main bottleneck of this method. It assumes you get each layer right before proceeding. In practice, revisions in layer one often require reworking layers two through four. The download situation. There isn't a single official software package called "Weather Architecture Jonathan Hill." The framework exists primarily as a set of methodologies and calculation approaches. Some components have been implemented in tools like ENERGY+. I found a set of spreadsheets and Excel-based calculators that someone translated from Hill's papers on a few architecture forums. These are community-shared, not officially endorsed, and the quality varies. The most useful one I found calculates insolation angles and thermal mass requirements from raw weather data, but it only handles temperate climates. If you are working in a tropical or arid zone, you need to adapt the parameters manually.

The main limitation of this approach is time. A complete Weather Architecture Jonathan Hill analysis for a single building typically takes two to three weeks for someone familiar with the method. A first-time user can expect three to four times that. The learning curve is steep because the methodology assumes you already understand basic building physics. If you don't, you'll spend most of your time looking up terms rather than applying the framework. I recommend pairing this with a simpler preliminary energy model first to build context. Another thing nobody warns you about: the weather data problem. Hill's framework requires high-quality hourly data. Many free weather databases provide daily maxima and minima or monthly averages. Neither is sufficient. You need hourly wind speed and direction, solar radiation, humidity, and temperature. Sites like Meteonorm or NASA's POWER database can provide this, but even these have gaps in older datasets. If your project is time-sensitive and the weather data isn't ready, the entire framework stalls. I've had projects delayed by six weeks waiting on interpolated weather data for a new construction site in an area without long-term meteorological records. Despite these issues, the method produces better results than conventional approaches for sites where climate is a primary design driver. I switched from using generic energy modeling to this framework about eighteen months ago and my passive heating and cooling strategies have been noticeably more effective on subsequent projects. The initial time investment pays off in the reduction of mechanical system requirements. On a recent project, specifying a smaller HVAC unit based on the Weather Architecture Jonathan Hill analysis saved roughly twelve percent on equipment costs compared to my previous projects using standard sizing methods. That number varies by region and building type, but it's a consistent pattern I've observed.

If you want to start with this, pull the available spreadsheets from architecture forums, get the best weather data you can for your site, and work through the layers one at a time. Don't skip the terrain correction step. Don't skip seasonal wind shifts. And don't expect to finish a full analysis in a week unless you have done this before. The first pass is always slower than you think.