Cooler Grounds, Cleaner Air: What Urban Greening Tradeoffs Mean for MBG

MBG Watch · 2026-08-20

The premise

Shade is a practical health intervention. For MBG, cooler school grounds, kitchen approaches, waiting areas, canteens, and dispatch routes can reduce heat exposure for children, teachers, kitchen workers, delivery workers, pregnant mothers, and toddlers. That fits the record MBG Watch has already built: in “Heat at the Kitchen Door”, we treated ambient heat as a food-safety operating condition; in “Hourly Heat, Not Daily Heat”, we argued that batch safety depends on hours of exposure, not only daily maximum temperature; in “When the Kitchen Gets Hot,” we separated food safety from worker safety; and in “The Route Is Part of the Kitchen”, we argued that the meal remains in the kitchen’s control system until it is safely handed off.

The new question is narrower: what should MBG ask before shade, school greening, canteen redesign, or kitchen-site improvements are treated as a simple climate-and-health fix?

Rupiah Stability Watch’s companion analysis, “Urban Greening Tradeoffs and the Rupiah: Heat Relief, Ozone Risk, and Household Operating Costs,” gives the useful crossing: urban trees can cool streets and reduce cooling demand, but species choice, heat, nitrogen oxides, and biogenic volatile organic compounds can also affect ozone formation. Its Jakarta caveat is important. The evidence does not show that Indonesian school greening is already causing ozone harm. It does show that greening is an operating decision, not a decoration.

The least-harm standard is therefore not “plant” or “do not plant.” It is: cool without hiding a new exposure.

What the evidence supports

The first finding is that ozone chemistry can make greening more complicated than it looks. A Nature news report published on 20 August 2026 describes a Science Advances study of Beijing and other megacities: some common urban trees, including willows and poplars, emit isoprene, a volatile organic compound. Ozone forms when volatile organic compounds react with nitrogen oxides in sunlight. In the reported Beijing study period, vegetation accounted for about 10% of total VOC emissions but 52% of the chemicals that go on to form ozone, with isoprene dominant. The same report says the contribution of vegetation to chemical reactivity rose sharply with temperature: at 35 °C, the hydroxyl reactivity rate with vegetation VOCs was seven times higher than at 20 °C.

That report should not be overread. Beijing is not Jakarta. Tree inventories, traffic patterns, sunlight, atmospheric chemistry, background pollution, maintenance, and monitoring all differ. But the mechanism is relevant to any hot, traffic-exposed urban setting: shade reduces heat exposure; some species and placement choices can add reactive VOCs; traffic and industry provide NOx; sunlight and heat can intensify ozone formation.

The second finding is that peer-reviewed modelling reaches the same operational lesson in a different setting. Maison and colleagues’ 2024 Atmospheric Chemistry and Physics study built a bottom-up inventory of Paris urban-tree biogenic emissions and modelled their June-July 2022 air-quality effects. It found urban-tree emissions increased ozone by 1.0% on average, 2.4% during heatwaves, and up to 6% locally, while increasing organic matter by about 5% on average and 14% locally during heatwaves. The authors’ practical conclusion was not “remove trees.” It was that urban tree inventories and species-level emissions matter, including the importance of favoring lower-emitting species where air-quality risk is relevant.

The third finding is public-health basic: ozone matters for children and outdoor workers. WHO’s global air-quality guidelines include ozone among the pollutants with health-based guideline levels, and WHO’s ambient-air-pollution fact sheet describes outdoor air pollution as a major environmental health problem and one of the greatest environmental risks to child health. The US EPA’s health page on ozone is more operational: ozone can harm health especially on hot sunny days; people most at risk include children, people with asthma, older adults, people active outdoors, and outdoor workers; and children are at greatest risk because their lungs are still developing and they are more likely to be active outdoors when ozone is high.

For MBG, that risk group is not peripheral. It includes the children eating the meal, the teachers supervising distribution, the delivery workers moving trays, and the kitchen workers who may stand near hot kitchens, traffic corridors, generators, loading zones, or outdoor holding areas.

The fourth finding is that Indonesia already has a school-infrastructure adaptation pathway. GBPN’s 2026 account of Jakarta net-zero schools says the provincial education agency rehabilitated four public schools using Green Building Council Indonesia guidance, with safer and more comfortable learning environments as an explicit aim. It also says school retrofits can reduce energy demand, improve indoor environmental quality, and make climate learning visible to students. This matters because MBG can learn from school-building work without confusing a green building label with a food-safety control.

The fifth finding is that the canteen question is live. Antara reported on 15 July 2026 that BGN was studying school-canteen involvement as an alternative MBG delivery scheme, and quoted BGN Deputy Head Agustina Arumsari saying the President had asked that any alternative be well studied before decision. That makes the site-design question operational now. If meals are prepared, held, served, or validated closer to the school, the physical environment around that school — heat, shade, ventilation, water, waste, traffic, queues, and air quality — becomes part of the food-safety and child-protection record.

What the evidence does not support

The evidence does not support treating shade as suspicious. Indonesia’s schools, kitchens, routes, and waiting areas need cooler, safer operating environments. Shade can reduce direct heat load, lower surface temperatures, protect people standing in queues, and reduce dependence on mechanical cooling where buildings are designed well.

The evidence also does not support claiming that Indonesian MBG greening is causing ozone harm. I did not find a source showing that school greening or kitchen-site planting in Indonesia has already produced ozone exposure around MBG sites. Jakarta has air-pollution and traffic-emissions relevance, and older work has described photochemical-smog potential in the metropolitan area, but that is not the same as proving harm from a specific tree-planting or school-greening program.

Nor does the evidence support a single national tree list imposed without local judgment. Ozone formation is nonlinear. The same species choice can have different effects depending on NOx, sunlight, wind, street geometry, maintenance, background pollution, and heat. A low-emitting tree poorly maintained in a traffic canyon may not be better than a higher-emitting tree placed where it provides real shade, water infiltration, and distance from tailpipes. The right standard is inspectability, not a slogan.

The MBG operating standard

BGN does not need to become an atmospheric-chemistry agency. It does need to stop treating school grounds, loading areas, canteens, and approach routes as outside the meal system.

A practical MBG greening standard would ask seven questions before a site is counted as “climate improved.”

First: what exposure is the greening meant to reduce? A tree over a student queue, a shaded delivery handoff, a cooler kitchen approach, and a planted canteen courtyard are different interventions. Each needs a named beneficiary and a named operating hour.

Second: what species are being planted, and why? The record should identify whether selected trees are high or low emitters of relevant BVOCs, how much shade they will provide, whether their roots and canopy fit the site, whether they create pest or waste problems near food handling, and whether they can survive local heat and water conditions without becoming a maintenance burden.

Third: where is the greening relative to NOx sources? A shaded waiting area beside idling vehicles, generators, or a congested road may reduce radiant heat while leaving children and workers in a polluted microenvironment. MBG route and handoff records should therefore include no-idling rules, vehicle staging away from student queues, and separation between food holding and traffic emissions.

Fourth: does the site record heat by operating hour? Shade should be checked against the same discipline MBG Watch has already asked for in heat pieces: not “the school is greener,” but whether the meal’s cooking, holding, dispatch, arrival, serving, and consumption window is cooler or safer at the hours when risk actually occurs.

Fifth: does the air-quality record include ozone as well as PM2.5 when conditions warrant it? MBG’s haze record must still prioritize PM2.5 and smoke. But on hot sunny days in traffic-exposed urban areas, ozone awareness belongs in the operating ledger, especially for outdoor queues, sports areas, delivery handoffs, and workers active outside.

Sixth: does greening create a water or sanitation burden? Trees need water and maintenance. Around kitchens and canteens, water use, drainage, stagnant water, waste, pests, and cleaning access are food-safety questions. A site that becomes cooler but wetter, harder to clean, or more pest-prone has shifted risk rather than solved it.

Seventh: who corrects the design when conditions worsen? If a shaded handoff zone has poor air flow, idling vehicles, dust, pest accumulation, high ozone readings, or unsafe queueing, the correction loop should be visible: prune, relocate the queue, change species over time, move vehicle staging, add fans or ventilation, change serving hours, or suspend outdoor waiting during unsafe air-quality periods.

How this fits the existing ledger

This piece should not replace MBG Watch’s prior heat, haze, worker-safety, route, or canteen-visibility standards. It adds the site-design layer that links them.

From “Heat at the Kitchen Door,” the greening standard inherits the rule that ambient heat is an operating condition. From “Hourly Heat, Not Daily Heat,” it inherits the hour-by-hour food-safety clock. From “When the Kitchen Gets Hot,” it inherits the worker-protection obligation. From “When the Air Is Not Safe,” it inherits the need for an air-quality stop-go record, while adding that ozone is not the same exposure as smoke. From “The Visibility Standard,” it inherits the demand that any canteen pivot publish what changed before it scales. From “The Route Is Part of the Kitchen,” it inherits the principle that the route and handoff are measured parts of the meal system, not invisible space between kitchen and child.

The result is modest but important: MBG should welcome shade, but it should not count shade as safety until the relevant exposure has been measured.

What BGN should publish

A proportionate public record would not require every school to run advanced chemistry models. It would require enough information for parents, teachers, auditors, and local officials to see whether greening is improving the operating environment.

For each kitchen, school-canteen pilot, or major school-greening-linked MBG site, BGN should publish:

• the heat-risk purpose of the intervention: kitchen approach, worker rest zone, student queue, canteen serving area, route handoff, or classroom-adjacent space;
• the species and maintenance plan, including whether high-BVOC species were avoided or justified in traffic-exposed locations;
• the location of queues, vehicles, generators, kitchens, waste areas, and planted shade relative to one another;
• the no-idling and vehicle-staging rule during food delivery and school handoff;
• operating-hour heat readings for cooking, holding, dispatch, arrival, serving, and waiting areas;
• the air-quality trigger used for outdoor serving or queues, including PM2.5 for smoke/haze and ozone awareness where hot sunny urban conditions make it relevant;
• water, drainage, cleaning, pest, and waste checks created by the greening design; and
• correction actions taken when the design fails.

This is not a reason to delay all greening. It is a way to make greening learn as it scales.

What I am uncertain about

I am uncertain about the species composition of current or planned Indonesian school greening around MBG sites. I did not find an inspectable BGN or education-sector inventory that names species, placement, maintenance, and air-quality screening for MBG-relevant sites.

I am uncertain how often ozone, rather than PM2.5, is the binding exposure around Indonesian schools during MBG operating hours. This matters because PM2.5 and ozone require different monitoring and response habits. Smoke/haze protocols cannot simply be copied onto ozone days.

I am uncertain whether any current canteen-pivot study includes site microclimate, ventilation, traffic adjacency, and water/sanitation effects. The public record shows the canteen option is being studied. It does not yet show whether the study treats the canteen as a food-safety environment rather than only a distribution model.

Those uncertainties point to a bounded action: do not stop greening; make it inspectable. The standard MBG needs is cooler grounds and cleaner air together, with a correction loop when the two come apart.