Wildfire risk: how to avoid unplanned downtime

In our experience with weather monitoring, wildfire is that threat that surrounds the operation from the outside: every business with vegetation next to its assets lives with it and, of course, so does our daily life.

There is wildfire monitoring by satellite and with information from neighboring regions. That matters. Being even more proactive means anticipating the conditions that can cause a fire and that are forecastable: dry air, weeks without rain, high temperatures, dry vegetation, and strong wind. The hazard, wildfire, has predictable triggers and, once they are diagnosed, it gains a probability of occurrence.

Probability of occurrence crossed with the impact on your sector, your business, your asset, and your operation becomes an actionable risk matrix: with a decision owner, contingency protocols, and clear actions to mitigate losses.

One sector is particularly affected and needs even more auditable measures: power transmission and distribution. The grid crosses the territory with vegetation on both sides, and the outage happens without the flame touching the wire.

Regional data gives the scale. Climate catastrophes cost Latin America US$ 11.6 billion in 2024, and only 13% of that value was covered by insurance; wildfire sits inside that count as part of the natural catastrophe aggregate. In Brazil, the region's largest market, about 65,000 power outages in 2024 started with fire near the grid, 38% more than the 47,000 occurrences of 2023 and more than double the 26,000 of 2020, according to a survey by Abradee with regulator data. 2024 was a peak and 2025 receded along with the burned area: MapBiomas recorded 12.7 million hectares burned in the country in 2025, about half of the previous year. The exposure remains the same: the same strip of vegetation next to the same wire, waiting for the next above-average dry season.

The bill does not stop at power. On July 23, 2026, a fire in the vegetation along highway BR-251, at km 472 in Francisco Sá (Minas Gerais, Brazil), cut visibility to the point that the road ran on half its lanes for about 20 minutes. Five thousand square meters of vegetation burned, with strong wind and steep terrain making the fight harder. With no damage reported to the pavement, the whole loss landed on the operation.

Without a timely warning of smoke over the road, soot under the wire, or fire in the right-of-way, wildfire collects on three fronts at once:

Life: whoever is in the path of the smoke or the flame. The driver entering a smoke curtain at 100 km/h, the maintenance crew scrambled at the last minute, the firefighter.
Operations: the asset stops before being damaged. The road closes, the line trips, the rail section is shut, generation drops under a clear sky.
Budget: every hour of that stoppage lands in the month's operating cost, on top of emergency repair and contractual or regulatory compensation.

Wildfire cuts across sectors that rarely talk to each other: the highway concessionaire, the rail operator, the power utility, the generation farm, and the insurer that covers them all. Each one records the same hazard through a different effect: road closure, train delays, outages, generation shortfall, property claims.

With the cost spread across different result lines, fire rarely shows up consolidated in a single report with the word wildfire under it. What shows up is the effect, already booked as something else.

In this article

The wildfire hazard: how it forms in Brazil, Chile, Mexico, and Europe.
How to turn the wildfire hazard into risk calculated by asset.
Wildfire monitoring and wildfire risk: detecting is not anticipating.
What the same wildfire impacts in each sector, starting with the table that sums it all up.
- Highways: the smoke closes the road before the fire reaches the asphalt.
- Railways: the right-of-way is the asset nobody accounts for.
- Power transmission and distribution: the outage happens without the flame touching the wire.
- Renewables: the smoke brings down generation under a clear sky.
- Insurance: exposure by asset improves wildfire pricing.
The return on investment from anticipating the hazard.
The method: three steps to apply starting tomorrow.
Where this becomes a daily decision.

The wildfire hazard: how it forms in Brazil, Chile, Mexico, and Europe

Fire needs three things at the same time: fuel, ignition, and an atmosphere that lets the fuel burn and the fire spread. Ignition, in most cases, is human. The fuel is the vegetation that is already there. The part that meteorology sees in advance, and that decides whether a hotspot dies within five meters or advances for kilometers, is the third.

That third part has four variables: heat, relative humidity, accumulated dryness in the fuel, and wind. Heat and dry air pull water out of vegetation in hours. The accumulated dryness of previous weeks defines how much fuel is already primed to burn. Wind carries the flame forward and the smoke far away. Operational indices such as the Canadian Fire Weather Index combine exactly these variables, and that is why the dangerous condition forms days before the first hotspot.

Mexico's national weather service publishes a daily bulletin with this reading, the Perspectiva Meteorológica para Incendios Forestales, and works with explicit thresholds: temperature from 35 °C, wind above 35 km/h, and relative humidity below 20% mark the combination with the highest risk of fire generation and spread.

Brazil: a long dry season with falling relative humidity. In July 2026, the Minas Gerais fire department responded to 2,334 fire occurrences, 48% more than June's 1,577, with afternoon relative humidity between 20% and 30%, highs near 32 °C, and 543 municipalities under alert. The same pattern repeats across the whole Cerrado between June and October, exactly the window in which the power grid, the rail right-of-way, and the highway margins are surrounded by dry fuel.

Chile: a strong anticyclone combined with a coastal low generates an easterly wind descending the Andes, the Puelche, which arrives hot and dry at the coast. In the Penco-Lirquén fire, in the Biobío region, between January 17 and 19, 2026, the Center for Climate and Resilience Research (CR2) measured 36.5 °C in Chillán on the 18th, relative humidity below 17%, and wind above 25 km/h during the initial spread. About 20,000 hectares burned in three days.

Mexico: a dry spring season with a high-pressure system over the center and west of the country. Conafor's spring 2026 assessment pointed to unusually high temperatures, low humidity, strong winds, and extreme dryness in forest fuels, with the State of Mexico and Jalisco among the hardest hit.

Europe: a prolonged heat wave over already-dry vegetation. In 2025, the European Union recorded 1,079,538 hectares burned, the highest value in the EFFIS series since 2006 and nearly double the 2006-2024 average. In the first three weeks of August, 22 very large fires in Portugal and Spain burned 460,585 hectares, 43% of the year's entire European total. World Weather Attribution measured a +4.6 °C anomaly over 16 days (August 3 to 18) in that episode and concluded that the observed fire weather was about 30% more intense and about 40 times more likely than it would have been without climate change.

Four countries, four local combinations, the same physical trigger: dry fuel, dry air, heat, and wind. That is why the wildfire hazard applies to any operation with vegetation nearby, in any of these markets.

How to turn the wildfire hazard into risk calculated by asset

Two operations cross the same dry season. The one that knows which of its sections enter critical condition, and in which week, clears, inspects, and positions crews beforehand. The other pays for improvisation. In practice, the difference shows up as clearing prioritized on the feeder that will concentrate risk, signaling inspection done on the right section before the closure, and a maintenance window moved in advance, with no wasted mobilization.

Wildfire is the hazard. Losses and risk to life are the risks. Between the hazard and the risks are factors that multiply or reduce the impact, and they decide whether the same critical condition burns five thousand square meters in an area with nobody around or closes a concession section at rush hour. i4sea follows three layers to make that conversion.

Layer 1, the data. The base is our proprietary numerical model with 1 to 3 km resolution across all of Latin America, calibrated with more than 10 years of real climate history. On top of that base run more than one hundred AI scenarios, used to understand the behavior of the weather that makes fire spread: which kilometers enter critical condition and which areas are most prone to trouble. Where public forecasting operates at roughly 25 km resolution and cannot tell your section from the rest of the region, this reading starts at the size of the concession kilometer, the feeder, the rail section, or the generation farm.

Layer 2, the interaction with territory and business. The same fire weather does not produce the same result at two different points. The reading crosses the hazard with:

Land use, vegetation cover, and slope: the real geography of that point. Steep terrain accelerates uphill spread.
Fuel load and condition along the strip: tall, dry vegetation with no recent clearing burns faster and hotter than a strip with upkeep in order.
Accumulated dryness: weeks without rain change the fire's behavior before any sign appears on the day of the event.
Wind direction relative to the asset: the same hotspot 300 meters from the line is irrelevant or critical depending on the wind direction.
What is in the smoke's path: fire in an area with no human or economic activity creates no business risk. The same smoke crossing a road, a line span, or a solar farm does.
What depends on the point struck: the more people, cargo, and revenue pass through that specific section, the higher the risk, even with an identical climate hazard.

Layer 3, the risk matrix. i4sea connects that threat's index, intensity, and hazard window at the exact point with the impact it causes in that specific business. That connection is what turns dry weather in the region into a risk reading calculated by asset, ready for decision.

These three layers sustain the four gains the operation feels: lower cost, less exposure of lives, crews allocated where the risk is, and planning closed before the event.

Wildfire monitoring and wildfire risk: detecting is not anticipating

Whoever searches for wildfire monitoring finds, in most results, satellite hotspot maps: where the fire is already burning right now. Whoever searches for wildfire risk finds public risk maps by region. Both layers work and have their place in the protocol. What neither answers is the question that defines the season's cost: which of my sections enter critical condition this week, and what does each crew do about it?

There are two clocks in play. The hotspot clock, of minutes to hours, belongs to satellite detection and firefighting, and serves to react to the fire that already exists. The window clock, of days to weeks, belongs to fire weather, and serves to clear, cut firebreaks, inspect, and position crews before the first hotspot. An operation working only with the first puts out fires; an operation working with both avoids the stoppage.

Complete weather monitoring adds the two layers plus a third: the reading by asset. A regional risk map says the week will be dangerous. The decision needs to know on which kilometer of strip, on which feeder, and in which farm the condition arrives first. The difference between the two readings is the difference between a generic alert and a work order.

What the same wildfire impacts in each sector

The physics is the same. What stops, which indicator suffers, and who signs off change completely.

Table 1 · What wildfire charges each sector

Sector	What fire and smoke stop	Where the bill shows up
Highways	Visibility on the road, right-of-way, signage	Road closure, accidents, concession performance indicator
Railways	Section shut, signaling cable and equipment	Downtime hours, cascade effect across the network
Power T&D	Line tripped by soot, feeder, easement strip	Continuity indicators, customer compensation, emergency repair
Renewables	Irradiance at the module, farm access, maintenance window	Generation shortfall, short-term market exposure, wasted mobilization
Insurance	Property claims on exposed linear assets	Pricing, provisioning, subrogation

Highways: the smoke closes the road before the fire reaches the asphalt

Start with the result. A concessionaire that knows on Monday which kilometers enter critical condition on Thursday sets up the detour with signage and communication planned, activates the variable message signs before the smoke curtain, and positions the field crew on the right section. The concessionaire that finds out on Thursday pays overtime, tow trucks, crew exposure, and user complaints.

Dense smoke over the road takes away the driver's visibility in seconds and, unlike fog, it forms in vegetation the concession itself manages.

The BR-251 event of July 23, 2026 shows the typical shape of the occurrence: five thousand square meters of vegetation, half the lanes for 20 minutes. What moves the performance indicator is recurrence within the dry season, and it is exactly that count that does not exist consolidated today.

The decision that changes place is clearing and firebreaks. A concessionaire with hundreds of kilometers under management does not clean everything in the same month, so it prioritizes. Today that prioritization tends to follow the calendar and complaint history. With wildfire risk calculated by segment, it starts to follow exposure: the kilometer with dry vegetation, favorable wind, and hotspot history goes first, before the season's peak, instead of waiting for the first closure.

The bill changes in three places. Planned closure: setting up a detour in advance costs signage and communication; setting it up in a scramble costs the rest. Firebreaks by real exposure: the same vegetation-management budget protects more critical kilometers when allocated by risk. A record for the granting authority: forecast risk, action taken, and results on file sustain the performance and rebalancing discussion with evidence.

Railways: the right-of-way is the asset nobody accounts for

On the highway, the driver detours. On the railway there is no alternative route, and the stopped section stops the entire operation behind it.

The operator that receives, days in advance, the information that three sections are entering high spread condition brings forward the inspection of those sections, positions firefighting resources near them, adjusts the scheduling of heavy trains, and warns the shipper before the cargo leaves. The operator without that reading discovers the problem with the track already shut and the train halfway there.

Wildfire attacks the railway on track, signaling, and traffic, all at once. Track: fire in the right-of-way shuts the section and exposes the maintenance crew. Signaling: heat and flame compromise cable and field equipment, which has long replacement lead times. Traffic: the whole network's schedule reorganizes in cascade, and that is where the cost multiplies.

To size the extreme of a prolonged shutdown, a 5-day blockage on the Vitória-Minas railway, in Brazil, generated R$ 645 million in losses. That blockage was caused by protest, not wildfire, and serves here only as an order of magnitude of what a stopped line represents for the sector. Divided by 120 hours of continuous operation, the value comes to about R$ 5.4 million per hour.

Anticipation protects, in this order: the crew that would go to the section unaware of the condition, the signaling equipment that prior inspection spares, and the relationship with the shipper, who takes a day-ahead warning far better than cargo stopped mid-route.

Power transmission and distribution: the outage happens without the flame touching the wire

On the railway, fire needs to reach the track. On the power grid, it needs to reach nothing. Smoke and soot under the line ionize the air, cause a short circuit, and trip the protection. The outage happens before any crew sees the fire, and the control center finds out through the loss of load.

A utility that sees wildfire risk by section chooses where clearing and inspection go first and where to position the on-call crew in the week the condition peaks. The cost of that choice is a clean strip. The cost of the wrong choice is interrupted load, customer compensation, and emergency repair.

The sector's numbers show why the choice weighs. On top of the roughly 65,000 national occurrences that open this article, Brazil's grid operator ONS recorded 210 disturbances on the core transmission grid caused by fires between January and August 2024, against 195 in all of the previous year, and wildfire became the second-largest cause of transmission interruptions, behind only adverse weather conditions, the category in which the sector groups phenomena such as lightning and windstorms.

In the state of Goiás, the cost comes from repetition. The local utility recorded 384 fire occurrences reaching the grid in 2025, affecting about 59,700 consumer units, concentrated between June and October. What weighs is the volume, with crews dispatched, customers without power, and continuity indicators under pressure at every episode.

Vegetation management answers where the risk strip is: registry, clearing cycle, inspection. Nobody answers in which week the spread condition peaks on that specific feeder. Without the temporal layer, clearing follows the annual calendar when it should follow risk, and the crew only mobilizes after the flood of calls.

Renewables: the smoke brings down generation under a clear sky

In transmission, fire takes the asset offline. In renewable generation, it attacks the resource before touching the equipment.

A farm manager who sees the spread condition by location days in advance moves the maintenance window to the week before, keeps the crew productive, and preserves availability. The manager who does not cancels on the day, pays for crew and crane mobilization without executing the work, and pushes maintenance to a worse window later.

In solar, dense smoke reduces the irradiance reaching the module. The sky is cloudless, the inverter is healthy, and the generation curve comes in below forecast. In weeks of widespread smoke, like the ones the 2024 fire season imposed on generators in Brazil's north and center-west, that gap between forecast and actual shows up directly in short-term market exposure, with no asset damaged.

In wind, fire in the farm's vegetation restricts crew access, cancels the scheduled maintenance window, and forces preventive turbine shutdowns for safety.

Insurance: exposure by asset improves wildfire pricing

The generation farm closes the side of whoever suffers the fire. The insurer looks at the same event from the other side of the table and reaches the same conclusion by another path.

Whoever reads exposure by asset, with history calibrated to the location, underwrites with less margin of error and negotiates provisioning better. Pricing wildfire exposure based on regional averages and claims history is operating with low-resolution information on a hazard that materializes at kilometer scale.

The asymmetry of the Brazilian market is measured. Insurance covers only 9% of the country's climate losses, against 20% to 55% in developed countries, and in the north and northeast coverage can fall below 2%. Of all claims paid in 2024, 58% came from the property line, exactly where substations, warehouses, yards, and exposed linear assets accumulate.

In claims adjustment, the gap is one of evidence. Without a record of what was known before the event and what was done with that information, the discussion about negligence, force majeure, and subrogation becomes a dispute of versions. A company that demonstrates active climate risk management, with the decisions taken before each event documented, arrives at renewal with evidence instead of narrative.

Calibrated underwriting: exposure by asset, not by municipality, reduces the asymmetry between the premium charged and the risk taken. Claims with evidence: the record of what was forecast, communicated, and executed sustains adjustment, subrogation, and technical provisioning. Dialogue with the insured: whoever manages risk has an objective argument at renewal, which changes the price conversation.

The return on investment from anticipating the hazard

No system prevents the dry season. What changes is the cost of crossing it.

The closest signal of wildfire impact comes from distribution itself. In Goiás, fire occurrences reaching the grid fell from about 700 in 2024 to just over 380 in 2025, a 44.8% drop the utility attributes to integrated preventive actions combined with improved weather conditions, in a year when Brazil's burned area fell by half. The number of affected customers, however, rose from about 49,000 to nearly 60,000 in the same period, up 21%. With fewer events and more customers affected per event, what remains to attack is the severity of each episode, and that is where the reading by section comes in.

The case below does not measure wildfire: it is here because it measures the mechanism this text defends, trading reaction for anticipation on critical assets, with an auditable number.

Vattenfall, wind power. An offshore wind farm in severe North Sea conditions reached an ROI of 27 times (2,749% of accumulated value) operating with hyperlocal 10-day forecasting per turbine instead of the global ECMWF model, with fewer unnecessary mobilizations and more maintenance windows used. The hazard was wind; the gain mechanism is the same: window decided beforehand, crew mobilized only once.

The logic behind the case has independent academic validation. An NBER working paper measured that the improvement in US hurricane forecast accuracy since 2007 cut the total cost of each hurricane by 23%, an average of US$ 2 billion per event. The hurricane kept happening. The quality of the preparation is what changed.

Applied to wildfire, the reasoning is identical: the fire happens, and the financial difference sits between the clean strip with the crew positioned on the right section and the emergency restoration with load interrupted.

The method: three steps to apply starting tomorrow

These steps work with a spreadsheet, an in-house system, or a vendor. None of them depends on hiring i4sea.

Register the sections, not the region. List the linear and point assets fire can stop: concession kilometers, rail sections, line spans, feeders, generation farms, warehouses, and yards. For each one, record three things: the fuel load along the strip, the sensitive load that depends on it, and the cost of one stopped hour there. Without that third piece, no prioritization survives a budget meeting, and that gap is internal to the operation.
Translate weather conditions into decision thresholds by asset. Define, for each section, the combination that triggers each risk level: accumulated dryness, relative humidity, temperature, wind, and hotspot history. The product of this step is one sentence per asset, with window, owner, and linked action, of the kind high risk on section X between Thursday and Saturday, prioritize inspection and position the crew. Also define who receives each warning: control center, maintenance, and safety need different things.
Record forecast, action, and result. Every time a risk level fires, save what was forecast, what action was taken, and what actually happened. Within one season you have your own hit rate, know which thresholds are poorly calibrated, and gain evidence for contract discussions, rebalancing claims, insurance notices, and board conversations.

The gain from the three steps is the same across the five sectors: the structurally lower cost of preparation replaces the cost of reaction. It is the Delta Framework i4sea uses in its benefit studies, and it fits in one line: benefit is the sum of the difference between unplanned cost and planned cost, multiplied by the number of events.

Where this becomes a daily decision

A wildfire risk threshold written into a procedure only protects the operation when the team receives the warning in time, on the channel it already uses, with the recommended action attached.

The AI Climate Agent delivers the decision to the team before the event: which section, which window, which action. It already knows your operation's thresholds, protocols, and history, arrives on the channel the team already uses, WhatsApp, Teams, or email, and records in every answer the source of the data behind it.

Try the Climate Agent on your operation, with a free 14-day trial and no credit card

If you prefer to start with a diagnosis, we map your assets' exposure to climate hazards and show you what the system would see today, with real data. Start with the most exposed section, not the whole network.

Request a free climate exposure diagnosis by asset →

The hours of warning your operation has before the next dry season, and what it does with them, decide the cost of the next fire.

Sources

Climate catastrophes cost Latin America US$ 11.6 billion in 2024, with only 13% insured. Swiss Re, sigma NatCat 2025. Scope: wildfire is not isolated; regional catastrophe aggregate.
About 65,000 fire occurrences cut power supply in Brazil in 2024, up 38% from 47,000 in 2023 and more than double the 26,000 of 2020 (Abradee, with regulator data). eixos, 2025 and G1, Jul 18, 2025.
ONS: 210 disturbances on core transmission lines caused by fires between January and August 2024, against 195 in all of 2023. MegaWhat, Sep 2024.
Wildfire as the second-largest cause of transmission interruptions, behind adverse weather conditions. Valor Econômico, Aug 31, 2024.
Equatorial Goiás: 384 fire occurrences reaching the grid in 2025, about 59,700 consumer units affected, critical months June to October. Jornal Opção, Jul 20, 2026. Operational data released by the utility itself.
Equatorial Goiás, 2024 vs 2025 series: about 700 occurrences in 2024 against just over 380 in 2025 (down 44.8%), with affected customers rising from about 49,000 to nearly 60,000 (up 21%). A Redação, Aug 4, 2026.
Mechanism of smoke and soot ionizing the air and causing short circuits on the grid. G1 Goiás, Sep 6, 2024.
BR-251, km 472, Francisco Sá (Minas Gerais, Brazil), Jul 23, 2026: vegetation fire cut visibility, traffic on half the lanes for about 20 minutes, about 5,000 m² burned. WebTerra, via O Tempo, Jul 24, 2026.
MapBiomas: 12.7 million hectares burned in Brazil in 2025, about half of the area burned in 2024 (25.4 million hectares). MapBiomas, Annual Fire Report, Jul 21, 2026.
Insurance covers only 9% of Brazil's climate losses, against 20% to 55% in developed countries, and below 2% in the north and northeast; 58% of claims paid in 2024 came from the property line. Radar de Eventos Climáticos e Seguros no Brasil, CNseg/EY, 2025. Official PDF.
Canadian Fire Weather Index. Canadian Forest Service / Natural Resources Canada. FWI use on the Iberian Peninsula: World Weather Attribution, 2025.
Mexico, official SMN thresholds: temperature from 35 °C, wind above 35 km/h, relative humidity below 20%; daily bulletin. Servicio Meteorológico Nacional / Conagua.
Brazil, Minas Gerais, July 2026: 2,334 fire occurrences attended by the fire department, up 48% from June's 1,577; afternoon relative humidity of 20% to 30%, highs near 32 °C, 543 municipalities under alert. Estado de Minas, Aug 7, 2026.
Chile, Penco-Lirquén fire (Biobío), Jan 17-19, 2026: about 20,000 hectares; 36.5 °C in Chillán on the 18th, relative humidity below 17%, wind above 25 km/h; synoptic context of a high-pressure ridge with a coastal low generating easterly wind (Puelche). CR2, Center for Climate and Resilience Research.
Mexico, Conafor spring 2026 assessment: unusually high temperatures, low humidity, strong winds, and extreme dryness in forest fuels, with the State of Mexico and Jalisco among the hardest hit. Tribuna de México, secondary outlet attributing the assessment to Conafor.
Europe, 2025 season: 1,079,538 hectares burned in the EU27, the highest in the EFFIS series since 2006; 22 very large fires in Portugal and Spain burned 460,585 hectares in three weeks of August, 43% of the EU total. Joint Research Centre, European Commission, Mar 31, 2026.
Europe, attribution of the August 2025 episode: +4.6 °C anomaly over 16 days; fire weather about 30% more intense and about 40 times more likely than without climate change. World Weather Attribution.
NBER: improvements in US hurricane forecast accuracy since 2007 cut total cost per hurricane by 23%, an average of US$ 2 billion per event. Molina, R. and Rudik, I., "The Social Value of Hurricane Forecasts", NBER Working Paper 32548 (2024). nber.org/papers/w32548.
R$ 645 million in losses over 5 days of blockage on the Vitória-Minas railway. i4sea reference base. Blockage caused by protest, not a climate event, and the body says so. The R$ 5.4 million per hour average is our own calculation (R$ 645,000,000 ÷ 120h, continuous operation assumption).
Offshore wind farm in the North Sea: 27x ROI (2,749% accumulated value), with hyperlocal 10-day forecasting per turbine instead of the global ECMWF model. Public Vattenfall case, compiled by i4sea. A wind case, not wildfire, and the body says so.
i4sea proprietary numerical model, 1 to 3 km resolution across Latin America, calibrated with more than 10 years of climate history; more than one hundred AI scenarios on top of the model; public forecasting at roughly 25 km. i4sea product documentation.
Combined reading of climate, asset, and territorial context; hazards monitored by sector, basis of Table 1; active climate risk management as a pricing argument; auditable record of forecast, time, owner, and action; Delta Framework. i4sea product documentation and technical material.
AI Climate Agent: free 14-day trial, no credit card, on WhatsApp, Teams, and email. [TO_CONFIRM: current commercial offer, 14-day trial or beta access.]
Volume and difficulty of the keywords "wildfire risk", "wildfire monitoring", and "fire weather": [VOLUME_TO_VALIDATE].
Landing page URLs for the Climate Agent and the free Diagnosis: [LANDING_AGENTE_CLIMATICO], [LANDING_DIAGNOSTICO] (placeholders, replace before publishing).