In January of last year, the Eaton and Palisades fires showed in clear and present terms the cost of inaction on the climate crisis. Fires are common in Southern California, a fixture really, yet January's fires were the worst in LA's history, pushed deep into the urban core by hurricane force winds.

Our cities operate in a twilight hour between two worlds. One world is the familiar machinery of departments, forms, and workflows that feel carved in stone. The other is the emerging landscape of adaptive, digitally native operations where information moves as fast as weather and where response can be shaped in real time.

That disconnect has been obvious and motivating for myself for roughly 1.42 decades (as well as many others for longer!). Saying that though is more akin to "Go West, Young Man," a loose direction rather than operational guidelines to upgrade our city operations technology stack.

One thing is certain though: the twilight will not last forever. What cannot endure will break, and the climate is already testing every brittle corner of our civic machinery. Fires, heat, and storms make that more clear every year. The only real question is whether we shape the transition or scramble after it.

My hometown of La Crescenta, a few miles from the Eaton fire burn scar, faces a familiar Southern California constellation of climate threats. Wildfire driven by wind and fuel loads. Flooding that quickly becomes debris flows and landslides. Earthquakes that test brittle infrastructure. Heat that stresses people, power systems, and water supply. And the cascading failures where one disruption amplifies another.

What makes La Crescenta interesting is not uniqueness but actually how profoundly middle of the bell curve it is for California suburbs. It's very close to the median size for California water retailers, for example. Profoundly ordinary. And what's increasingly important for profoundly ordinary places is the integration layer, the connective tissue that enables more effective city operations.

Digital twins provide a useful aspiration for that integration layer. Not the overhyped smart city-esque marketing technoblather, but the OG University of Michigan vision of a useful bridge between virtual and real space. The Digital Twin Consortium has a useful definition of the term:

A data-driven virtual representation of real-world entities and processes, synchronized at a specific frequency and fidelity

The key word there is "synchronized." A digital twin is not a static map or a dashboard. It's a feedback loop. A representation that updates as reality changes and, in turn, shapes decisions that change reality.

But here's the thing. That feedback loop already exists. It's just not digital, not integrated, and often not visible.

Every suburb in Southern California already runs on protocols. Not the TCP/IP kind (though those too), but the operational kind: the scheduled routines, inspection cadences, trigger thresholds, and coordination sequences that keep a place functioning. A street sweeper follows a route on a weekly cycle. See debris basin inspections and brush clearance mandates. Water mains get replaced on a schedule (currently accelerated as the system was built out in the 50s and 60s and now is at the end of its useful life).

These are feedback loops between URL and IRL, between the virtual world of planning, scheduling, regulation, and modeling and the physical reality of pipes, basins, trees, and fire. A water quality sampling schedule is a protocol: it defines what gets measured, how often, where, and what triggers a response. A red flag warning is a protocol: atmospheric conditions cross a threshold, a warning propagates through institutional channels, and pre-positioned resources activate. Brush clearance enforcement is a protocol: a regulatory deadline triggers inspection, inspection produces compliance data, noncompliance triggers enforcement action.

The protocols stack. They operate at different temporal scales, from the SCADA system polling water pressure every fifteen seconds to the regional transportation plan projecting decades out. Each layer has its own rhythm, its own institutions, its own failure modes. And the gaps between layers, the places where a fast protocol and a slow protocol fail to coordinate, are where cascading failures live.

A digital twin, understood properly, is not a replacement for these protocols. It's a way to level them up. To increase the frequency and fidelity of the feedback loops that already exist. To make visible the coordination gaps between layers. To compress the latency between observation and response.

Here's what that protocol stack looks like for a suburban Southern California community like La Crescenta:

That last question is the one that stays with me. Every suburb in California was designed for a climate that is departing. The protocols that maintain these places, from hourly dispatch to decadal planning, were calibrated to a world that is changing faster than the institutional cycles that govern response. The digital twin aspiration is really a question about whether we can recalibrate those feedback loops before the gap between assumption and reality becomes catastrophic.

Throughout my professional career and curious passion for city operations, the gap between what theoretically should be done per global best practices, academic theory, and other intellectually correct goals and what happens on the ground has always struck me.

Human error, resource constraints, coordination headwinds... many real world factors mean that often obvious needs like having digital geographic information system (GIS) layers for basic infrastructure don't always exist or aren't integrated and shared across the relevant utilities. Basic infrastructure in the US has layers upon layers of rules, procedures, administration, insider institutional politics that prevent familiar obstacles to urban digital tool applications.

I have found that there is no substitute for getting into the field to see firsthand what is actually going on and use that analogue data as a tool for designing what might be.

Being a data sophist of sorts, I have a recurring daydream to design a digital twin armed with pencil, paper and a few print books like the handy MIT guide to "Street Fighting Mathematics." Arguably with sufficient completeness, a janky contraption toy model composed of coordinates, flows and stocks is a bridge between the virtual and real world. Just synced at a relatively low fidelity and high latency.

My young family recently had our second kid. When my first daughter was young, I loved wearing and walking her everywhere. As I'm taking some bonding leave, I'm excited at the opportunity for a little urban exploration and protocol watching, to rediscover the small town in Southern California where we live.

The protocol stack above is a field guide of sorts. At every temporal scale, there are things you can see if you know what to look for. The spatial distribution of rooftop solar, both at a neighborhood-wide scale and in terms of rooftop orientation. The logic of fire hydrants, storm drains, power lines, flood control channels, debris basins, natural gas lines, sidewalk placement, bus lines. The dispersion of trees: geographically, by apparent age, by species. The traffic flow at peak up and down Briggs Terrace, a neighborhood with one ingress and egress.

Each observation is a data point in a very low fidelity digital twin. A notebook full of coordinates, flows, and stocks. The protocol stack gives that notebook structure, a way to sort observations by the temporal rhythm they belong to and the institutional layer they reveal.

There's a story about William Mulholland, LA's famous water engineer who served as the de facto founder of the modern city, that has stayed with me. Before a single pipe was laid for the Los Angeles Aqueduct, he walked the route. Two hundred and thirty miles through the Mojave Desert and the Sierra Pelona Mountains, bacon and beans for meals, sleeping on frozen ground, barometer in hand.

The barometer was the key instrument: by measuring air pressure at each step, Mulholland was calculating altitude, hunting for the precise grade that would keep water moving under gravity alone, fast enough to flow but slow enough not to tear the concrete apart. Too steep and the channel erodes. Not steep enough and the water stalls. The whole system depended on finding a continuous, calibrated descent across some of the most punishing terrain in North America. By the time construction began, he claimed he knew every foot of the route.

I plan to keep that story with me when I wander around La Crescenta with a little human. You cannot understand a place from a dashboard. You cannot model what you haven't seen. A walk is a protocol observation instrument. A notebook is a model. The friction between what you imagine a place to be and what it actually is, that gap, that surprise, is where the real work begins.

Where does the built environment assume a climate that no longer exists? That's the question I'll carry.

In January of last year, the Eaton and Palisades fires showed in clear and present terms the cost of inaction on the climate crisis. Fires are common in Southern California, a fixture really, yet January's fires were the worst in LA's history, pushed deep into the urban core by hurricane force winds.

Our cities operate in a twilight hour between two worlds. One world is the familiar machinery of departments, forms, and workflows that feel carved in stone. The other is the emerging landscape of adaptive, digitally native operations where information moves as fast as weather and where response can be shaped in real time.

That disconnect has been obvious and motivating for myself for roughly 1.42 decades (as well as many others for longer!). Saying that though is more akin to "Go West, Young Man," a loose direction rather than operational guidelines to upgrade our city operations technology stack.

One thing is certain though: the twilight will not last forever. What cannot endure will break, and the climate is already testing every brittle corner of our civic machinery. Fires, heat, and storms make that more clear every year. The only real question is whether we shape the transition or scramble after it.

My hometown of La Crescenta, a few miles from the Eaton fire burn scar, faces a familiar Southern California constellation of climate threats. Wildfire driven by wind and fuel loads. Flooding that quickly becomes debris flows and landslides. Earthquakes that test brittle infrastructure. Heat that stresses people, power systems, and water supply. And the cascading failures where one disruption amplifies another.

What makes La Crescenta interesting is not uniqueness but actually how profoundly middle of the bell curve it is for California suburbs. It's very close to the median size for California water retailers, for example. Profoundly ordinary. And what's increasingly important for profoundly ordinary places is the integration layer, the connective tissue that enables more effective city operations.

Digital twins provide a useful aspiration for that integration layer. Not the overhyped smart city-esque marketing technoblather, but the OG University of Michigan vision of a useful bridge between virtual and real space. The Digital Twin Consortium has a useful definition of the term:

A data-driven virtual representation of real-world entities and processes, synchronized at a specific frequency and fidelity

The key word there is "synchronized." A digital twin is not a static map or a dashboard. It's a feedback loop. A representation that updates as reality changes and, in turn, shapes decisions that change reality.

But here's the thing. That feedback loop already exists. It's just not digital, not integrated, and often not visible.

Every suburb in Southern California already runs on protocols. Not the TCP/IP kind (though those too), but the operational kind: the scheduled routines, inspection cadences, trigger thresholds, and coordination sequences that keep a place functioning. A street sweeper follows a route on a weekly cycle. See debris basin inspections and brush clearance mandates. Water mains get replaced on a schedule (currently accelerated as the system was built out in the 50s and 60s and now is at the end of its useful life).

These are feedback loops between URL and IRL, between the virtual world of planning, scheduling, regulation, and modeling and the physical reality of pipes, basins, trees, and fire. A water quality sampling schedule is a protocol: it defines what gets measured, how often, where, and what triggers a response. A red flag warning is a protocol: atmospheric conditions cross a threshold, a warning propagates through institutional channels, and pre-positioned resources activate. Brush clearance enforcement is a protocol: a regulatory deadline triggers inspection, inspection produces compliance data, noncompliance triggers enforcement action.

The protocols stack. They operate at different temporal scales, from the SCADA system polling water pressure every fifteen seconds to the regional transportation plan projecting decades out. Each layer has its own rhythm, its own institutions, its own failure modes. And the gaps between layers, the places where a fast protocol and a slow protocol fail to coordinate, are where cascading failures live.

A digital twin, understood properly, is not a replacement for these protocols. It's a way to level them up. To increase the frequency and fidelity of the feedback loops that already exist. To make visible the coordination gaps between layers. To compress the latency between observation and response.

Here's what that protocol stack looks like for a suburban Southern California community like La Crescenta:

That last question is the one that stays with me. Every suburb in California was designed for a climate that is departing. The protocols that maintain these places, from hourly dispatch to decadal planning, were calibrated to a world that is changing faster than the institutional cycles that govern response. The digital twin aspiration is really a question about whether we can recalibrate those feedback loops before the gap between assumption and reality becomes catastrophic.

Throughout my professional career and curious passion for city operations, the gap between what theoretically should be done per global best practices, academic theory, and other intellectually correct goals and what happens on the ground has always struck me.

Human error, resource constraints, coordination headwinds... many real world factors mean that often obvious needs like having digital geographic information system (GIS) layers for basic infrastructure don't always exist or aren't integrated and shared across the relevant utilities. Basic infrastructure in the US has layers upon layers of rules, procedures, administration, insider institutional politics that prevent familiar obstacles to urban digital tool applications.

I have found that there is no substitute for getting into the field to see firsthand what is actually going on and use that analogue data as a tool for designing what might be.

Being a data sophist of sorts, I have a recurring daydream to design a digital twin armed with pencil, paper and a few print books like the handy MIT guide to "Street Fighting Mathematics." Arguably with sufficient completeness, a janky contraption toy model composed of coordinates, flows and stocks is a bridge between the virtual and real world. Just synced at a relatively low fidelity and high latency.

My young family recently had our second kid. When my first daughter was young, I loved wearing and walking her everywhere. As I'm taking some bonding leave, I'm excited at the opportunity for a little urban exploration and protocol watching, to rediscover the small town in Southern California where we live.

The protocol stack above is a field guide of sorts. At every temporal scale, there are things you can see if you know what to look for. The spatial distribution of rooftop solar, both at a neighborhood-wide scale and in terms of rooftop orientation. The logic of fire hydrants, storm drains, power lines, flood control channels, debris basins, natural gas lines, sidewalk placement, bus lines. The dispersion of trees: geographically, by apparent age, by species. The traffic flow at peak up and down Briggs Terrace, a neighborhood with one ingress and egress.

Each observation is a data point in a very low fidelity digital twin. A notebook full of coordinates, flows, and stocks. The protocol stack gives that notebook structure, a way to sort observations by the temporal rhythm they belong to and the institutional layer they reveal.

There's a story about William Mulholland, LA's famous water engineer who served as the de facto founder of the modern city, that has stayed with me. Before a single pipe was laid for the Los Angeles Aqueduct, he walked the route. Two hundred and thirty miles through the Mojave Desert and the Sierra Pelona Mountains, bacon and beans for meals, sleeping on frozen ground, barometer in hand.

The barometer was the key instrument: by measuring air pressure at each step, Mulholland was calculating altitude, hunting for the precise grade that would keep water moving under gravity alone, fast enough to flow but slow enough not to tear the concrete apart. Too steep and the channel erodes. Not steep enough and the water stalls. The whole system depended on finding a continuous, calibrated descent across some of the most punishing terrain in North America. By the time construction began, he claimed he knew every foot of the route.

I plan to keep that story with me when I wander around La Crescenta with a little human. You cannot understand a place from a dashboard. You cannot model what you haven't seen. A walk is a protocol observation instrument. A notebook is a model. The friction between what you imagine a place to be and what it actually is, that gap, that surprise, is where the real work begins.

Where does the built environment assume a climate that no longer exists? That's the question I'll carry.