[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"blog-traffic-noise-analysis-beta":3},{"id":4,"title":5,"author":6,"authorAvatar":7,"authorEmail":8,"authorLink":9,"body":10,"category":317,"description":318,"extension":319,"gradient":320,"mainImage":321,"meta":322,"navigation":336,"path":337,"pinned":338,"publishedAt":339,"readTime":340,"seo":341,"stem":342,"__hash__":343},"blog\u002Fblog\u002Ftraffic-noise-analysis-beta.md","Traffic Noise in Spacio: find the quiet side while the massing still moves","Franz Forsberg","\u002Fimg\u002Favatar_franz.webp","franz@spacio.ai","https:\u002F\u002Fwww.instagram.com\u002Ffranz.forsberg\u002F",{"type":11,"value":12,"toc":305},"minimark",[13,18,26,29,36,43,47,54,57,64,70,79,83,89,95,105,111,121,127,133,139,149,155,159,166,169,173,180,186,193,197,208,215,222,226,229,247,251,256,277,282,293,296],[14,15,17],"h2",{"id":16},"the-report-that-arrives-after-the-massing","The report that arrives after the massing",[19,20,21,22],"p",{},"On most housing projects near a busy road, the noise question is answered by a report. The acoustic consultant models the scheme, the report comes back, and by then the massing has been agreed with the client and drawn up for the planning meeting. If the report says the corner flats have no quiet side, the answer is glazing, ventilation and a floor plan redrawn around the problem. ",[23,24,25],"strong",{},"The massing stays where it was, because moving it now costs weeks.",[19,27,28],{},"I worked for years as a building performance analyst in Norway, and noise was the analysis that most reliably arrived too late to change anything big. Not because anyone ignored it. Because nobody in the design team had a number until the report landed, and an argument without a number loses to one with a number.",[19,30,31,32,35],{},"Today we are releasing ",[23,33,34],{},"Traffic Noise in Spacio, in beta",". It puts that number in the room while the massing still moves.",[19,37,38],{},[39,40],"img",{"alt":41,"src":42},"A courtyard block in Traffic Noise, every facade and roof in the 5 dB bands, with the verdict: every building has a quiet side","\u002Fimg\u002Fblog\u002Ftraffic-noise-analysis-beta\u002Foverview.webp",[14,44,46],{"id":45},"it-answers-the-design-question-not-how-loud","It answers the design question, not \"how loud\"",[19,48,49,50,53],{},"A noise map on its own tells you how loud a site is. That is rarely what you have to decide. What you have to decide is ",[23,51,52],{},"whether a facade can take a bedroom, whether each building has a quiet side to put the bedrooms on, and whether a window can stay open at night",". So that is what the tool leads with.",[19,55,56],{},"After a run, the panel opens with a sentence rather than a percentage: whether the buildings have a quiet side, how much of the facade is quiet enough to plan a bedroom against, and how much of it can keep a window open at night. Below it sit the rule it was read against, what the run covered, and a chart of how much facade and outdoor area falls in each 5 dB band.",[19,58,59,60,63],{},"The rule comes from the project's own country: Norway's T-1442, Sweden's traffic noise ordinance (written in Leq over 24 hours rather than Lden, and read that way), Denmark's Miljøstyrelsen guidance, and the WHO guideline of 53 dB Lden everywhere else, rather than a neighbour's regulation. The night reading is always held against ",[23,61,62],{},"WHO's 45 dB night guideline",", whatever the national rule, because the question it answers is whether someone can sleep with the window open.",[19,65,66],{},[39,67],{"alt":68,"src":69},"Plan view in the night period: receivers along every facade, the verdict and the band chart in the panel","\u002Fimg\u002Fblog\u002Ftraffic-noise-analysis-beta\u002Fverdict.webp",[19,71,72,73,78],{},"If you want the longer argument for why noise belongs in the first week of a project, and the health case behind it, it is in ",[74,75,77],"a",{"href":76},"\u002Fblog\u002Ftraffic-noise-cities-architects-design","Traffic noise in cities: what architects can actually do about it",". This post is about the tool.",[14,80,82],{"id":81},"a-run-from-placing-the-disc-to-trying-a-move","A run, from placing the disc to trying a move",[19,84,85,88],{},[23,86,87],{},"Place the study area."," Click to place a disc over your buildings, with a 150 m radius by default. While you drag it, Spacio lights what the run will read: the part of each plot inside the disc, and the facades and roofs of every building at least half inside it. The streets are coloured by their speed, with a legend on the canvas, so you see the sources before you start.",[19,90,91],{},[39,92],{"alt":93,"src":94},"Placing the study area: the disc over the block, the buildings the run will read lit, the streets coloured by speed","\u002Fimg\u002Fblog\u002Ftraffic-noise-analysis-beta\u002Fplacement.webp",[19,96,97,100,101,104],{},[23,98,99],{},"Check the streets."," Every street becomes a line of sound sources, from its type, its speed and its traffic. Where you have typed a speed, a daily volume or a heavy-vehicle share on a street, the run uses yours. In Norway, where you have not, it uses the count from the national road database, NVDB: daily traffic, long-vehicle share and posted limit. Otherwise it takes the default for the street type. ",[23,102,103],{},"The run tells you how many streets were counted and how many you set by hand",", because a map of this street and a map of a street like this one are different claims. Railways and tram lines are sources too, each running a train mix you can name and change.",[19,106,107,110],{},[23,108,109],{},"Press Start and read it."," Levels are read on a grid over the ground, on every facade at every storey, and on the roofs. The colours are the 5 dB bands official European noise maps use, fixed from 35 to 85 dB, with yellow starting at 55 and red at 65: the two lines T-1442 draws. Because the scale is fixed, two runs compare with each other, and a run compares with an official map.",[19,112,113,116,117,120],{},[23,114,115],{},"Ask a point why it is loud."," Press Alt and hover a facade. Spacio draws the paths the sound takes to that point: straight, over the roofs, around the corners, off the facades opposite. It lists the streets that make it loud, largest first, until together they carry what would have to go to meet the limit. ",[23,118,119],{},"That list is often the most useful thing on screen."," If one street carries most of it, a screen or a turned slab on that side will do the work. If it takes five streets, no single road is the problem; the layout is.",[19,122,123],{},[39,124],{"alt":125,"src":126},"Inspect mode: the traced paths to one point in the courtyard, 42.6 dB Lden, and the street that makes it loud","\u002Fimg\u002Fblog\u002Ftraffic-noise-analysis-beta\u002Finspect.webp",[19,128,129,132],{},[23,130,131],{},"Switch to Spaces."," Here the map gives way to your project: every space coloured by how exposed it is, on one ladder. Green is fine. Light green has a quiet side, so plan the bedrooms there. Amber needs a fix, and red needs a different plan. You can ask three questions of it: the national rule; Sleep, whether a bedroom can keep a window open at night; and Daytime, the hours offices, schools and living rooms are in use. Homes, schools, care and hotels are judged on their quiet side, other uses on the glazing they would need. Hover a space to see why.",[19,134,135],{},[39,136],{"alt":137,"src":138},"The Spaces view: 72 of 179 sensitive spaces need a fix, each space coloured by its quiet side","\u002Fimg\u002Fblog\u002Ftraffic-noise-analysis-beta\u002Fspaces.webp",[19,140,141,144,145,148],{},[23,142,143],{},"Try a move."," Close the gap in the perimeter block, raise the edge along the road, turn the slab so its gable faces the street. The run follows, and ",[23,146,147],{},"the verdict says what changed, leading with a quiet side gained or lost",". A change to traffic alone, a speed or a volume, re-reads the paths already traced instead of tracing them again; on the project we timed it on, that took about a third of a second.",[19,150,151],{},[39,152],{"alt":153,"src":154},"A perimeter block closed on all four sides: the courtyard facades and roofs fall into the quiet bands","\u002Fimg\u002Fblog\u002Ftraffic-noise-analysis-beta\u002Fmove.webp",[14,156,158],{"id":157},"the-levers-and-what-each-one-is-honest-about","The levers, and what each one is honest about",[19,160,161,162,165],{},"The moves that matter most are drawn: continuity and height along the loud edge, courtyards that close, slabs that turn their short end to the road, a noise-tolerant bar that takes the exposure for the housing behind it. The tool does not quote a rule of thumb for those. ",[23,163,164],{},"It shows what each one is worth on your site",", which is the point of having it.",[19,167,168],{},"Four more are strategies, switches the result can be read with: a speed limit, soft ground, low-noise asphalt and absorptive facades. They are not things you draw, and the panel is plain about that. A speed limit is a policy lever you argue for with the municipality. Soft ground reads the whole site as planted, so it is the upper bound of what planting can do. Low-noise asphalt is laid by the road owner and loses some of its effect over the years. Absorptive facades take in the first bounce between buildings and into courtyards, and do nothing for the facade that faces the road directly. While any strategy is on, the result says so, so a number read with a 30 km\u002Fh cap never gets quoted as the scheme as drawn.",[14,170,172],{"id":171},"held-against-the-official-map","Held against the official map",[19,174,175,176,179],{},"Where a country publishes its strategic noise map as data, Spacio fetches it and drapes it on the terrain in the same colours, with your run's facades kept on top so one can be read against the other. Then it compares, receiver by receiver: ",[23,177,178],{},"the share of the ground inside the official map's band, above it and below it, and the mean difference in decibels",". Today that covers Norway (Statens vegvesen's road map, Bane NOR's railway map and the T-1442 planning zones), Helsinki, Brussels and Stockholm.",[19,181,182],{},[39,183],{"alt":184,"src":185},"Statens vegvesen's strategic noise map draped on the terrain, with the run's facades on top","\u002Fimg\u002Fblog\u002Ftraffic-noise-analysis-beta\u002Fofficial-map.webp",[19,187,188,189,192],{},"On Filipstad in Oslo, the run read ",[23,190,191],{},"0.4 dB above Statens vegvesen's map on average",", with 45% of the ground inside its band, 39% above and 15% below. Two things to keep in mind when you read a comparison like that. The official map is itself a model, so this is one model held against another; the Norwegian road map is computed from 2021 traffic and the railway map from 2016. And an official map is drawn 4 m above the ground, where Spacio reads at 1.5 m by default, the height outdoor limits for gardens, balconies and play areas are written at. Set the ground reading to 4 m and the comparison is like for like; the readout always says which height it read.",[14,194,196],{"id":195},"how-far-to-trust-it","How far to trust it",[19,198,199,200,203,204,207],{},"The method is ",[23,201,202],{},"CNOSSOS-EU",", the common assessment method the EU set out in Directive 2015\u002F996, with the emission coefficients as corrected in 2021. Our implementation passes the ISO\u002FTR 17534-4 reference cases TC01 to TC07, and agrees with NoiseModelling 6.0.0, the open-source reference implementation, to 0.02 dB on average over a shared test scene. On a handful of real projects we have put a Spacio run next to the acoustic consultant's own modelled report for the same site, and ",[23,205,206],{},"the levels were typically within 3 dB of theirs",". That is roughly the range competent models agree within. None of it is a comparison against field measurements, because we have not made one.",[19,209,210,211,214],{},"It is also ",[23,212,213],{},"not a compliance check",". It does not produce a T-1442 report, and it does not model the night maximum level. It reads road and rail, not aircraft or industry. Counted traffic is Norway only for now. It runs on your graphics card, so it needs a browser with WebGPU, and it warns you on a machine with under 4 GB of graphics memory.",[19,216,217,218,221],{},"Which brings me to the position I care about most. Traffic Noise is for the early phase, where today there is no acoustic feedback at all and the massing gets decided without it. ",[23,219,220],{},"It does not replace your acoustic consultant, and in beta that is more true, not less",": the consultant is what quality-assures the project. What changes is what you bring to the first meeting with them. Better massing, a quiet side on every building that needs one, and fewer surprises, because the cheap decibels were spent on geometry before anyone had to buy the expensive ones in glazing.",[14,223,225],{"id":224},"try-it-on-the-site-you-are-working-on","Try it on the site you are working on",[19,227,228],{},"Open a project near a road, place the study area, and look at the corner flats first. Then close the gap and run it again.",[19,230,231,241,242,240],{},[23,232,233,234,240],{},"Traffic Noise is in beta at ",[74,235,239],{"href":236,"rel":237},"https:\u002F\u002Fapp.spacio.ai",[238],"nofollow","app.spacio.ai","."," Tell us where it is wrong as well as where it is right, in ",[74,243,246],{"href":244,"rel":245},"https:\u002F\u002Fdiscord.gg\u002FjdAGrfUBbY",[238],"Discord",[14,248,250],{"id":249},"sources","Sources",[19,252,253],{},[23,254,255],{},"Regulations and standards",[19,257,258,259,262,263,265,266,268,269,271,272,274,275],{},"⁃ Commission Directive (EU) 2015\u002F996 establishing common noise assessment methods (CNOSSOS-EU), and Commission Delegated Directive (EU) 2021\u002F1226 amending its emission coefficients.",[260,261],"br",{},"\n⁃ ISO\u002FTR 17534-4:2020, Acoustics, software for the calculation of sound outdoors, part 4: recommended quality requirements and test cases (TC01 to TC07).",[260,264],{},"\n⁃ T-1442\u002F2021, Retningslinje for behandling av støy i arealplanlegging, Klima- og miljødepartementet (Norway): 55 dB Lden yellow zone, 65 dB Lden red zone. The night maximum level is a separate requirement and is not modelled.",[260,267],{},"\n⁃ Förordning (2015:216) om trafikbuller vid bostadsbyggnader (Sweden), written in Leq over 24 hours.",[260,270],{},"\n⁃ Miljøstyrelsen guidance on road traffic noise (Denmark).",[260,273],{},"\n⁃ World Health Organization, Environmental Noise Guidelines for the European Region, 2018: road traffic noise below 53 dB Lden and below 45 dB Lnight.",[260,276],{},[19,278,279],{},[23,280,281],{},"Data and reference implementations",[19,283,284,285,287,288,290,291],{},"⁃ NVDB (Nasjonal vegdatabank), Statens vegvesen: daily traffic, long-vehicle share and posted speed, Norway.",[260,286],{},"\n⁃ Official strategic noise maps used for comparison, never presented as Spacio's result: Statens vegvesen road map (END round 4, 2021 traffic) and T-1442 støyvarselkart; Bane NOR railway map via Miljødirektoratet (END round 3, 2016 traffic); City of Helsinki 2022; Bruxelles Environnement 2021; Stockholm Bullerkartan 2022 (Leq over 24 hours).",[260,289],{},"\n⁃ NoiseModelling 6.0.0, open-source CNOSSOS-EU implementation, used as the cross-model reference: 0.02 dB bias and 0.04 dB spread over 77 receivers on a shared courtyard scene.",[260,292],{},[294,295],"hr",{},[297,298,299],"blockquote",{},[19,300,301],{},[302,303,304],"em",{},"Franz Forsberg studied architecture and engineering in Japan and worked as a building performance analyst in Scandinavia. He took an MSc at the Architectural Association (AA) in London, where his thesis, \"The Potential use of Natural Ventilation in Office Buildings in Tokyo\", set the through-line of his work: overheating, thermal comfort and natural ventilation treated as questions of massing and form rather than late-stage fixes. In 2022 he co-founded Spacio with André Agi and Stian Haugrim, with the aim of making climate and comfort design accessible to every architect, not just the specialists.",{"title":306,"searchDepth":307,"depth":307,"links":308},"",2,[309,310,311,312,313,314,315,316],{"id":16,"depth":307,"text":17},{"id":45,"depth":307,"text":46},{"id":81,"depth":307,"text":82},{"id":157,"depth":307,"text":158},{"id":171,"depth":307,"text":172},{"id":195,"depth":307,"text":196},{"id":224,"depth":307,"text":225},{"id":249,"depth":307,"text":250},"Product","Traffic Noise is in beta. It tells you where bedrooms, open windows and quiet sides work on your site, while the massing is still moving.","md","linear-gradient(135deg, #1A1F36 0%, #2D3561 100%)","\u002Fimg\u002Fblog\u002Ftraffic-noise-analysis-beta\u002Fhero.webp",{"slug":323,"pillar":324,"tags":325,"status":330,"sources":331},"traffic-noise-analysis-beta","Core Product",[326,327,328,329],"Noise","Urban climate","Site & context","Feasibility & Massing","Draft",[332,333,334,335],"https:\u002F\u002Feur-lex.europa.eu\u002Flegal-content\u002FEN\u002FTXT\u002F?uri=CELEX:32015L0996","https:\u002F\u002Feur-lex.europa.eu\u002Flegal-content\u002FEN\u002FTXT\u002F?uri=CELEX:32021L1226","https:\u002F\u002Fwww.who.int\u002Feurope\u002Fpublications\u002Fi\u002Fitem\u002F9789289053563","https:\u002F\u002Fnoise-planet.org\u002Fnoisemodelling.html",true,"\u002Fblog\u002Ftraffic-noise-analysis-beta",false,"2026-09-30","11 min read",{"title":5,"description":318},"blog\u002Ftraffic-noise-analysis-beta","RvlROz5SfHkiQwebfGghObBUspVEb9RkV1_ic9in8Qo"]