Sustainable Singapore

Data

Several Lines of Thought

When I began collecting data (or rather, asked Claude Code to implement the data as a live tracker in this page), I discovered two different records. Since I'm a history guy, I wanted to go as far back as 1900, but only Berkeley Earth's data was available (and goes all the way back to the nineteenth century).[^3] The Berkeley Earth record also ends at 2020, which wouldn't suit my purpose as I wanted a living app that constantly updates every year.

Fortunately, the data from Changi's climate station, which is the official temperature record used by Singapore, is available online on data.gov.sg. However, this same national record only begins in 1982,[^2] which leaves a gap of 82 years. So I'm combining both records to fill in for each other. That said, you'll notice a difference of 0.4 °C between them in the years they overlap. Apparently, Berkeley Earth's record is based on a regional field average, whereas Changi climate station's is much more specific.

Loading data…

The station doesn't remain stationary

Singapore's climate station has constantly moved over the decades, and as you can see from the historical record above, even the nation's land is continuously developing and building. Consequently, the temperature measured will inevitably be entwined with local construction.

A manhwa-styled illustration of a young man at a desk at night, frowning at a tablet with printed pages spread in front of him

Heatmap

Hold up, let Singapore cook

While NEA stations measure air temperature of specific sites in Singapore, this Landsat collection focuses on surface temperature. That is to say, the temperature of the ground, such as tarmac, roofs, or leaf canopy. Surface readings are usually higher than air readings, which is why the numbers are much higher than the 28 °C annual mean temperature in the timeline chart above. However, they are still important because they display the heat distribution across Singapore, and as you study it, you'll be able to notice a pattern.

The heatmap is built from thirteen cloud-free Landsat 8 and 9 passes over Singapore between April 2022 and April 2026, each taken at about 11:16 a.m. Each pixel is 30 meters across.[^39]

Satellite land-surface temperature across Singapore. The airports and the
western industrial estates are the hottest ground on the island; the central
catchment forest is the coolest.

Changi Airport47.6 °C

Jurong Island45.9 °C

Orchard40.5 °C

Lim Chu Kang36.8 °C

Central catchment33.5 °C

33 °C 47 °C and above no clear reading

The surface temperatures here are based on the median of thirteen Landsat overpasses.[^39] There are limitations to this data, though. Indicated in gray are areas that the passes were unable to capture clearly, perhaps obstructed by clouds, or regions that fall outside the satellite range, such as the northeastern corner, which includes Pulau Ubin and Pulau Tekong.

What we can see for sure is that Changi Airport's runways are the hottest surface in the country at 47.6 °C, with Jurong Island not far behind at 45.9 °C. This ties in with what we know, the island being reclaimed land, paved and roofed from one end to the other, and home to much of Singapore's petrochemical industry. Orchard Road also boils at 40.5 °C, 3.7 °C hotter than Lim Chu Kang, which is close to the 4 °C gap measured between them at street level in Roth’s study, despite being only 30 minutes apart.[^23]

Unsurprisingly, the coolest surface in Singapore is the central catchment forest at 33.5 °C, 14.1 °C lower than Changi Airport's runways, and among the very few regions yet to be urbanized.

Reconciling the two maps

If you notice any discrepancies between NEA station's live map and the Landsat-based heatmap, such as why the former's afternoon measurement of Orchard Road sometimes inclines toward the coolest in Singapore, it's because they're measuring two different things. Air and land. Furthermore, NEA climate stations are usually located in open, ventilated ground, far away from other buildings. This makes their thermometers unlikely to measure temperatures captured within a street surrounded by high-rise buildings, or an urban environment. In contrast, satellite readings simply measure whatever surface they capture.

A manhwa-styled illustration seen over the shoulder of a young man typing at a laptop, a night window and city lights beyond the desk

How many years at each temperature
Annual mean	Earlier	Later

The data between 1900 and 1981 was taken from Berkeley Earth,[^3] while the figures from 1982 onward are retrieved from Changi Climate Station.[^2] There is a discrepancy of 0.35 °C between them, which means the numbers for 1900–1962 appear lower. However, it is clear that the top 11 range (27.4 °C to 28.4 °C) are all clustered from 1963 onward. To be more precise, every reading in the top 11 range actually comes from 1982 or later, as Singapore became increasingly urbanized.

Synthesis

My read of the data

From 26.7 °C in 1900, the temperature has climbed to 28.1 °C in 2025, after reaching a height of 28.4 °C the year before. Of note, though, is when Singapore began its rapid urbanization in 1960, when HDB was first formed and began constructing housing for the population. Remarkably, the mean temperature remained consistent during that decade, constantly hovering around 26.7 °C. There are two sharp increases to take note of, such as 26.9 °C in 1966, which if you recall, was when the second five-year program of the Toa Payoh reclamation began, and 27.1 °C in 1969, which was when it neared its completion before people moved in on 1970, which was also a high 26.9 °C.

Aside from 1941, 1958 and 1969, where temperatures hit 27.2, 27.1 and 27.1 °C respectively, Singapore's annual mean never went above 27 °C. That changed from the mid-1960s onward, when temperatures consistently hovered around the 27 °C mark. Interestingly, the temperatures remained consistent throughout the 1960s and first half of the 1970s (though you will notice a clear increase from 1966 onward), despite Singapore's reclamation and rapid urbanization efforts during that period. I wonder if the heat might have been mitigated by Greening Singapore, the Garden City program and the annual Tree Planting Day in 1963, 1967 and 1971, with these early green initiatives bearing some fruit.

By 1977, which was well into the expansion of both Changi Airport and Jurong Industrial Estate, Singapore's annual mean hit 27 °C — and on the Changi record, which begins in 1982, it has dropped below that mark only twice since (in 1984, when Godzilla returned, and 1989, when Godzilla fought Biollante). As you can see from the table alongside this section, the annual mean temperature of Singapore only hit 27.4 °C and above from 1963 onward, or more accurately, if you refer to the temperature chart above, from 1982 or later. That's also when Tuas was industrialized, Changi Airport was considerably expanded, the merger that became Jurong Island began, and the East Coast reclamation entered its final phases. Before 1982 and the mass urbanization of Singapore, the annual mean temperature never exceeded 27.2 °C, based on Berkeley Earth. The efforts of Greening Singapore, the Garden City program and the annual Tree Planting Day might have been outstripped by the rapid pace of urbanization, the planting of trees no longer able to keep up with the proliferation of high-rise buildings that replaced much of Singapore's wilderness.

From the 1990s, temperatures began increasing sharply once more, and though the 1997 high of 28.3 °C is often attributed to the El Niño event that wreaked havoc across the world and continued into 1998, I'm not sure the surge is completely unrelated to Jurong Island's reclamation. Indeed, around 2002, when Jurong Island was on the verge of completion and petrol and fossil fuel companies began occupying it, the temperature shot up to 28.1 °C and slightly dropped to 27.8 °C after its official opening.

In fact, after the 2000s, the temperature has regularly hit 28 °C or above, about 1.1 °C warmer than in 1960. Never mind 1900, over a century ago, in about 60 years, we went from constantly below 27 °C to frequently above 28 °C. If you feel like Singapore has gotten hotter recently, it's not your imagination. Between 2008 and 2021, the number of HDB units grew by 24%,[^16] and the land area increased by 22.9 km².[^5] And that's just HDB alone, without counting the construction of private estates (e.g. condominiums) and industrial facilities. During this period, several condominium clusters were constructed in Pasir Ris Grove, such as NV Residences in 2013, The Palette in 2015, D'Nest in 2017, Coco Palms in 2018, and even the upcoming Pasir Ris 8 that connects directly with Pasir Ris Mall and MRT interchange. Further complexes, such as Seastrand in 2014, Ripple Bay in 2015, and Sea Horizon in 2016, were raised within the same neighborhood. The apparent condominium boom appears to have very tangible effects on the climate. Accordingly, the annual mean temperature rose by 0.43 °C. The years hitting 28 °C and above doubled from 4 between 1982 and 2007 to 8 in the next 18 years.

The heat is real.

Projection

The future is looking hot

The annual mean temperatures might appear to fluctuate at first glance, but upon closer examination, you'll notice a single consistent trend. A trend made visible if you were to extrapolate it in a single red line. Undoubtedly, temperatures in Singapore have been consistently rising. Even with fluctuations, they'll rarely diverge from the shaded band. In other words, by 2035, the annual mean temperature may reach as high as 29 °C while being unlikely to dip below 27.8 °C.

—

warming per decade, on the Changi record[^2]

—

the projected temperature ten years later

—

the range the temperature will most likely fall under

The next century's climate

If we are to extrapolate the line toward the next century, 2100, you'll see that temperatures are projected to hit 29.9 °C then. They could be as high as 32.9 °C, or perhaps hover around 28.5, depending on Meteorological Service Singapore's scenarios, but either way reflect a worrying trend if nothing major changes. Do take note, though, that this is simply a projection based on today's +0.22 °C per decade trend, which we then compared against the range taken from MSS's Third National Climate Change Study. That study runs climate models under three emission scenarios, their projections based on a 1995–2014 baseline.[^34] This is not a confident prediction of the future, but just a calculation of where we are headed based on contemporary data. Our calculations fall in the middle scenario of MSS's, based on our unbending adherence to a straight line throughout.

Tree angles to consider

Throwing shade

Of course, Singapore isn't sitting around doing nothing as our little island nation heats up to uncomfortable levels. Urban planners have conducted research, which led them to believe that shading roads with trees can reduce street-level temperatures by 2.6 °C. Furthermore, with buildings under tree shade, air conditioning demand will lessen.[^25]

Lei Xu, Ronita Bardhan, Hao Mei, Srilalitha Gopalakrishnan, Xing Zheng and Thomas Schroepfer's study revealed that street shade does indeed provide effective cooling, with a decrease of 3.1 °C in Universal Thermal Climate Index, 0.8 °C in air temperature, 10.0 °C in mean radiant temperature, and a significant reduction in heat stress severity. They also mentioned other studies where natural shading from trees has been observed to reduce the Universal Thermal Climate Index by up to 10.3 °C, while overhead shading devices lower it by 5.3 °C.[^26] Similarly, Puay Yok Tan, Nyuk Hien Wong, Chun Liang Tan, Steve Kardinal Jusuf, Mei Fen Chang and Zhi Quan Chiam note that vegetation canopy's temperature reduction composes of 29% evaporation cooling and 71% shade, with an average decrease of 1.7 °C.[^29] Do take note that their measurements were based on potted saplings instead of trees. In other words, tree shade is an effective way of mitigating heat in urban Singapore.

Naika Meili, Xing Zheng, Yuya Takane, Ko Nakajima, Kazuki Yamaguchi, Dengkai Chi, Yue Zhu, Jing Wang, Yeshan Qiu, Athanasios Paschalis, Gabriele Manoli, Paolo Burlando, Puay Yok Tan and Simone Fatichi noted that trees provided a reduction of air conditioner energy consumption of 8% in Singapore, and up to 12% in Tokyo during summer. However, they also observed that simply planting more and more trees to get more coverage might not necessarily yield better effects. The optimal cover is 40% tree cover, and anything beyond that will lead to a smaller air conditioner energy consumption reduction.[^27]

Lai Fern Genevieve Ow and Eugenie Chan observed that the rain tree, with its larger leaf density, crown area, sap wood area, vessels and rays, was superior to the yellow flame in providing shade and lowering air temperatures beneath the canopy, making pedestrians feel less hot outdoors.[^28] If you recall, the rain tree was the sapling Dr. Goh Keng Swee planted on Mount Faber on the first Tree Planting Day in 1971. Even before that, the rain tree was among several chosen for their rapid growth and reputations as “instant trees” when the Garden City campaign was implemented in 1967, which also included the angsana, sea apple and curtain creeper.[^20] Even today, it remains a reliable choice in guarding Singapore against climate change.

However, not all scholars agree. Matthias Roth, for example, said that research has not validated the success of green initiatives, such as planting more trees and developing rooftop gardens on buildings, in reducing the Urban Heat Island effect.[^23] Nonetheless, green initiatives should go beyond rooftop gardens and planting trees, and also encompass a wider range of heat mitigation measures, such as reducing use of fossil fuels, developing renewable energy sources and increasing energy efficiency. After all, decreasing heat emission should be just as much a priority as planting more trees for heat absorption.

Michael Alonzo, Peter Ibsen and Dexter Locke's review included a study that observed how dry regions were better off favoring shade over the moisture released by trees during transpiration, and even then, increased tree canopy was only effective at cooling the air in hotter and drier Southern Europe. Worse, in extreme, tropical humid conditions like Singapore, it appeared that increased tree canopy might have only a negligible impact on cooling, though they acknowledged more studies are required in this climate zone.[^32] In other words, while preserving our greenery is a priority, we should not be overly optimistic about its immediate effects.

A manhwa-styled illustration of a boulevard roofed over by the meeting canopies of mature rain trees

Forest area Loading…

This chart covers the forest area in Singapore, measured by the Food and Agriculture Organization of the United Nations (FAO) and published by the World Bank.[^45] Take note that FAO only has data for 1990, 2000, 2010, 2015 and 2020. The other years are simply a rough interpolation, and not an actual measurement. Contrary to expectations, it appears that Singapore’s forest area increased from 148.3 km² in 1990 to 170.1 km² in 2000 before peaking at 177.4 km² in 2010. However, given the decline of forest area to 164.7 km² in 2015 and 155.7 km² in 2020, I don’t think it’s a coincidence that this is within the same period where the annual mean temperature rose by 0.43 °C, and that 2016 and 2019 were the hottest years on record (28.4 °C). As forest area fell from 2010, four years within that decade reached 28 °C or above, equaling the four that hit the same threshold between 1982 and 2007. And we haven’t counted the data on 2021 and after, given that we have no actual measurements of forest area then.

Happy Tree Friends

The cool trees on the block

So what would be the best trees to plant to help mitigate rising temperatures? Here’s a list of trees that are great at providing comfort from the heat.

A manhwa-styled illustration of a rain tree — **Rain tree** *Samanea saman* With large leaf density, crown area, sapwood area, vessels and rays, the aptly named rain tree provides significant shade and reduces air temperature under its canopy.[^28] If you recall, the now ubiquitous rain tree was also one of the “instant trees” planted during the Garden City campaign in 1967.

A manhwa-styled illustration of a trumpet tree — **Trumpet tree** *Tabebuia rosea* As a seasonally dry forest tree, it has the potential to grow rapidly and have high rates of transpiration in a humid environment such as Singapore. With higher stomatal conductance and wider vessels in the xylem, larger potential for sap-flow, its higher transpiration rate will lead to greater evaporative cooling potential.[^47]

A manhwa-styled illustration of a pride of india — **Pride of India** *Lagerstroemia speciosa* Belonging to the seasonally dry forest group, like the trumpet tree, it has a higher transpiration rate during cloudy conditions, and thus greater evaporative cooling potential.

A manhwa-styled illustration of a flame of the forest — **Flame of the forest** *Delonix regia* Also belonging to the seasonally dry forest group, like the trumpet tree and pride of India, it is included here because of its potentially high transpiration rate and evaporative cooling. Furthermore, it is more tolerant of drought than evergreen species, such as yellow flame.[^47]

A manhwa-styled illustration of a Brazilian ironwood — **Brazilian ironwood** *Caesalpinia ferrea* Belonging to the seasonally dry forest group, it displayed high transpiration and heat transfer rates under sunny conditions, which makes it good for cooling in Singapore on such cloudless days.[^47]

A manhwa-styled illustration of a North Indian rosewood — **North Indian rosewood** *Dalbergia sissoo* Part of the seasonally dry forest group, it also has high transpiration and heat transfer rates under sunny conditions, which also makes it great for providing residents relief from heat when there are no clouds.[^47]

What transpired during my research

Despite Tan et al.’s hypothesis that higher transpiration rates have larger evaporative cooling potential in their research comparing seasonally dry forests against aseasonal evergreen forests,[^47] other studies observed that this might not necessarily be the case. For example, Ow and Chan’s comparison between the rain tree and yellow flame (Peltophorum pterocarpum) revealed that while the latter has a higher transpiration rate, it was less effective at cooling than the former,[^28] because shade was the greater contributor to reducing heat (71% shade, 29% transpiration, see above).[^29]

This is supported by Li et al., who found that in a tropical place like Singapore, the island’s high humidity reduces the cooling efficacy of urban trees’ transpiration. In fact, they found that in countries with a humid climate, excessively planted trees or those planted in enclosed spaces would result in low cooling efficacy or thermal discomfort, and might even raise the perceived temperature by up to 2 °C.[^48]

If you’ve been searching around local websites like me, you might have stumbled upon a webpage, Green Garden, which claimed that “native trees like Rain Tree, Angsana, Sea Apple, and Tembusu are perfect” for providing shade and cooling Singapore.[^49] But my own research couldn’t find evidence to prove that. In the first place, the rain tree is not native to Singapore. It is believed to be native to South America, specifically Colombia, the Caribbean slope and Orinoco drainage of Venezuela, as well as Central America.[^50]

Teo, Yu and Luo compared the tembusu against eight other trees in Singapore, and found it was among the most temperature-sensitive, its photosynthetic capacity falling by nearly 50% over a 5 °C increase.[^51] What does photosynthesis have to do with cooling, you may ask. Well, it’s directly related to transpiration. So if the tembusu photosynthesizes less, it transpires less. This is because leaves open the same stomata to absorb carbon dioxide and release water vapor. So if fewer of the tembusu’s stomata are open when photosynthesizing, it also means less water vapor is released. And when transpiration falls, so does its cooling efficiency.

Compared to the rain tree and mahogany in Bandung, Indonesia, the angsana had the lowest cooling efficiency, reducing air temperature by only 2.28 °C. For reference, the rain tree scored 3.8 °C and the mahogany 3.1 °C. The angsana also had the lowest humidity increase under its canopy (again, the rain tree scored the highest), which also speaks about a lower transpiration rate. The angsana has a higher average canopy area than mahogany, though, with around 223.55 m² compared to the latter’s 108.45. Once again, the rain tree triumphs with a 513.44 m² average canopy area.[^53] Not exactly confidence-instilling for the angsana. As for sea apples, they are, in fact, being replaced by NParks because they have been observed to be susceptible to basal rot.[^52] So they’re being removed before they start falling over and becoming a hazard to residents.

A shady simulation

The trees have us covered...or do they?

Based on the data Claude and I gathered, we built a mathematical model that calculates how much air conditioner energy we can save in proportion to tree cover. As mentioned in Meili, et al's findings, once you have more than 40% cover, the air conditioner energy savings begin to slow drastically. Apparently, that many trees will add so much humidity it begins to tax the air conditioners instead.

Tree cover of the street

Tree height, against the buildings

The street itself

−16%

summer air-conditioning energy, against the same street with no trees at all

What do these numbers mean? Meili, et al obtained data for Singapore at 40% cover, which they then averaged across every type of street. This translates to 5.92 million people, living in roughly 15 m² homes each, saving 2.39 kWh per m² per year, assuming their electric bill costs SG$0.32 and 0.41 kg of CO₂ per kWh. This totals SG$68 million and 87,000 tons of CO₂ a year, and that's before taking into consideration the CO₂ these trees will absorb. Meili, et al simulated five different cover percentages (0%, 20%, 40%, 60% and 80%), but they acknowledge that their ventilation test might affect results. E.g. a building with lots of ventilation halves the energy savings, whereas a sealed structure gains about 25% more savings. They also calculated based on the average across a whole street, rather than a single tree outside a specific building.[^27]

In contrast, Chen Yang, Mengju Xie and Thushini Mendis's review calculated that energy savings in equatorial climates can reach up to 31.75%, which is almost double Meili, et al's data, but they measure the shade provided by a single tree planted three to five meters away from a wall.[^31] Shamila Haddad's team, studying Riyadh with a different model at a different university, only calculated 13.4% energy savings from irrigated tree cover across 3,323 buildings in the Al Masiaf district,[^33] instead of the 17% obtained by Meili's team for the same city. Do take note these values differ from the simulator because it calculates Singapore's streets, which average 8%, and not 17%.

On the other hand, Alonzo, Ibsen and Locke corroborate the 40% cover, their own studies placing the optimal cooling within a canopy threshold of 35 to 50%.[^32] The takeaway, then, is yes, planting more trees will help mitigate the Urban Heat Island Effect, but only to a certain extent. The solution isn't as simple as planting more trees. We need to plant more trees, ensure we're not excessive about it (or the extra humidity might create new problems), and consider other options.

After all, combating climate change isn't growing greenery to absorb the heat alone. I am not naive enough to believe we should simply stop urbanization, nor am I delusional enough to demand we demolish buildings and replace concrete with jungle. We have already gone past the point of no return, and like it or not, urbanization is inevitable, and more buildings will be constructed in Singapore as we progress into the future. That is the nature of civilization. The government might acknowledge the climate crisis, but they also have to consider the needs of the population, whether housing, financial well-being (in the form of employment), health and infrastructure (both digital and industrial). There is no going back to some idealized, romanticized world of nature free of industry and technology. Rather, I suggest that even as we continue to urbanize, we take measures to mitigate the climate changes that arise with it, whether in the form of developing more efficient energy systems, using nature-friendly materials, pioneering environmental engineering techniques or combining urbanization with reforestation. That is to say, ensuring there is abundant space for planting trees even in new, modern urban districts.

We also need to cut down on carbon emissions, whether it is switching to clean energy or electric cars. However, even if we use electric cars, the power we use shouldn't be generated from burning gas, otherwise we're back to square one. That means looking into how we could produce our energy.

The nuclear option

Running low on gas

In 2024, Singapore's fuel mix for electricity generation consisted of natural gas (94%), solar (2.1%), coal and petroleum (1.2%) and other energy products, such as municipal waste, biomass and imports (2.7%). The percentage of natural gas dropped to 93.1% in the first half of 2025, while solar rose to 2.5%.[^35] Electricity is generated by burning gas, which leads to high carbon emission, resulting in rising temperatures.

Given my history with studying Japanese history (that was on purpose), which included Godzilla, the atomic bombs in World War II, the proliferation of nuclear reactors across postwar Japan, and eventually the 2011 Fukushima disaster, I can't say I'm an advocate of nuclear technology. However, I am less of a fan of relying on fossil fuels for power, especially when confronted with the challenge of climate change, so I thought I should cautiously consider calculating the option. Regardless of my personal reticence, Singapore is looking into nuclear energy. Current Prime Minister Lawrence Wong said in his Budget 2025 speech in February that year that the government will study the potential deployment of nuclear power in Singapore. Later that year, in September, the Energy Market Authority commissioned Mott MacDonald to conduct a safety and technical feasibility study on advanced nuclear energy technologies, such as small modular reactors.[^37] On 1 March 2026, EMA signed a Memorandum of Understanding with Korea Hydro and Nuclear Power Co Ltd on civil nuclear energy cooperation, as part of exploring how Singapore can decarbonize our power system and ensure reliable, secure and sustainable energy supply.[^38]

Simulated percentage of power supply provided by reactors

—

million tons of CO₂ a year that stop being emitted[^35]

The difference in heat boils down to efficiency. While nuclear reactors are capable of generating far more power than their gas counterparts, their thermal efficiency is 33–37% in light water reactors, though the latest pressurized water reactors can hit 38%.[^36] In contrast, our current gas-powered electricity generation is roughly 50%, with a gross output of 60 TWh from a 123 TWh input in 2024, and 30 TWh from 60 TWh in the first half of 2025. Also note that in 2024, the carbon emission stood at 0.402 kg CO₂ per kWh.[^35]

Heated absence

Instead of calculating temperatures (trust me, I tried, but Claude Code advised against it), this model is framed around carbon emissions instead. We don't have the data to calculate how much impact switching to nuclear energy would have on Urban Heat Island effect or Singapore's annual mean temperature. Furthermore, the excess in waste heat emitted by nuclear reactors would point the other way, anyway. However, we do have data on how much carbon emissions nuclear energy can reduce, and given that decarbonization is a global effort and that carbon dioxide, a greenhouse gas, is one of the major causes of rising temperatures, I feel tackling the root of climate change is worth exploring.

Takeaways

Too long, didn’t read?

If you skipped to this section because you found me longwinded, and wanted me to get straight to the point, well, here’s the short version.

A manhwa-styled illustration of HDB blocks along a street under a bright midday sun

Singapore is hot!

Ever since Singapore began reclaiming and large-scale urbanization in the 1960s, the annual mean temperatures have been regularly hovering near 27 °C,[^3] and it started hitting 27.4 °C and above from 1982,[^2] when our country’s modernization went into full swing.

A manhwa-styled illustration of an avenue of broad-crowned trees casting dappled shade

Planting a seed of skepticism

While most professionals would recommend planting more trees to reduce urban heat, and they are correct, there is a caveat. Firstly, there is little cooling benefit to exceeding 40% tree cover, as the humidity created from excessive transpiration may backfire.[^27] Secondly, despite the longstanding species that have taken root in Singapore since the 1960s, by virtue of Greening Singapore, the Garden City program and Tree Planting Day, some trees are better at cooling than the others.[^47] Worse, a couple of those species even ended up proving susceptible to basal rot.[^52]

Looking at you, rain tree.[^28]

A manhwa-styled illustration of a small modular reactor vessel in an industrial hall

Reacting to small modular reactors

Though there is an advantage in switching from fossil fuel-based electricity to nuclear power, the greatest of which is a stark reduction in carbon emissions (and thus contributing to the war against climate change),[^35] the larger waste heat produced means that even as we contribute globally to protecting the climate, Singapore doesn’t get cooled as effectively.[^36]

Character reference of Trevor, the protagonist of Reincarnator X Regressor: a young man in black glasses and a red school blazer, drawn in manhwa style

Disclaimer

The artwork isn't real

To avoid getting into copyright trouble, I decided not to copy historical photographs from National Archives of Singapore, National Library Board, etc., and instead generated the illustrations using my self-built webcomic MCP ecosystem. I used three of my MCP servers: webcomic background generator, character and panel generator, and anime production generator (which created the video you see above, albeit not very well done).

As such, my illustrations should be read as interpretations of documented backgrounds, and not an accurate visual representation of historical reality.

Also, I thought it was a good chance to put something I built with my buddy, Claude Code, to use, and demonstrate the fruits of our webcomic labor.

Footnotes

References

Where relevant, I have left Claude's notes as they are, because they surprisingly provide valuable insights on how to read the data. This is not to say Claude did all the research. I sourced most of them. Plenty of Claude's online sources were dead links or locked behind paywalls, so I had to do some manual digging. Of course, I read them myself (even if I didn't quite understand the more technical and scientific articles). I only fed them to Claude afterward to give him context and to ensure my read of the articles was right (especially when interpreting numerical data), not to ask him to read for me. Nonetheless, I think this is evidence that AI and humans can work well together.

Climate and Land Data

Meteorological Service Singapore. “Past Climate Trends.” Accessed August 14, 2026.
National Environment Agency. “Surface Air Temperature.” Data.gov.sg. Five datasets, Changi Climate Station, January 1982 onward: “Monthly Mean”; “Annual Mean Daily Maximum”; “Annual Mean Daily Minimum”; “Monthly Absolute Extreme Maximum”; “Monthly Absolute Extreme Minimum.” Accessed August 14, 2026. The annual mean and the annual hottest and coldest readings are derived here from the monthly series, complete twelve-month years only; MSS does not publish them. 1997 and 2015 both land on exactly 28.25 °C, so this page shows 28.3 where MSS’s table shows 28.2. The ten warmest years are the same either way.
Berkeley Earth. “Singapore: Regional Average Temperature.” singapore-TAVG-Trend.txt. https://berkeleyearth.org/data/. Accessed August 14, 2026.
National Environment Agency. “Real-Time Air Temperature API.” Data.gov.sg. https://api-open.data.gov.sg/v2/real-time/api/air-temperature. Accessed August 14, 2026.
Singapore Department of Statistics and Singapore Land Authority. “Land Area (As at December), Annual.” Data.gov.sg. Accessed August 14, 2026.

Housing, Reclamation and Industry

National Library Board. “Housing and Development Board.” Singapore Infopedia. Accessed August 14, 2026.
National Heritage Board. “The First HDB Blocks & the HDB Terraces.” My Queenstown Heritage Trail. Roots.gov.sg. Accessed August 14, 2026.
National Library Board. “Bukit Ho Swee Fire.” Singapore Infopedia. Accessed August 14, 2026.
Housing & Development Board. “Toa Payoh.” Our Towns and Estates. Accessed August 14, 2026.
National Heritage Board. “Toa Payoh Heritage Trail.” Roots.gov.sg. Accessed August 14, 2026.
National Library Board. “Toa Payoh.” Singapore Infopedia. Accessed August 14, 2026.
“Toa Payoh in the 1960s.” Remembering HDB Estates (blog), April 2018.
Lim, Tin Seng. “Land from Sand: Singapore’s Reclamation Story.” BiblioAsia 13, no. 1 (2017).
National Heritage Board. “Jurong New Town and Jurong Industrial Estate.” Roots.gov.sg. Accessed August 14, 2026.
National Library Board. “Marina Barrage.” Singapore Infopedia. Accessed August 14, 2026.
Housing & Development Board. “Dwelling Units under HDB’s Management, by Town and Flat Type.” Data.gov.sg. Accessed August 16, 2026. Analysis by the author, not a citation. Counts are as at 31 March of each financial year and cover sold and rental units together. Summed across every town and flat type, the stock under HDB’s management goes from 883,896 units in FY2008 to 1,096,380 in FY2021 — a rise of 24.0%.

Environment and Greening

National Heritage Board. “Rainforests by the Sea: Celebrating Singapore’s Mangrove Forests.” Roots.gov.sg. Accessed August 14, 2026.
National Parks Board. “Greening with the Community.” Annual Report 2023/24.
Thulaja, Naidu Ratnala. “Tree Planting Campaign.” Singapore Infopedia. National Library Board. Accessed August 14, 2026.
Lim, Tin Seng. “The Blossoming of Tree Planting Day.” BiblioAsia 17, no. 4 (2022).
National Parks Board. “OneMillionTrees Movement.” Trees.sg. Accessed August 14, 2026. The count is a live counter on that page. It read 887,044 in August 2026 and will have moved since.
National Parks Board. “More than Half a Million Trees Planted with the Community.” News release, April 15, 2023.

Urban Heat Island and Tree Cooling

“Killer Heat.” Sustainable Regional Network, NUS Faculty of Arts and Social Sciences, January 3, 2019.
Priyadarsini, Rajagopalan, Wong Nyuk Hien, and Cheong Kok Wai David. “Microclimatic Modeling of the Urban Thermal Environment of Singapore to Mitigate Urban Heat Island.” Solar Energy 82, no. 8 (2008): 727–745.
“Research Cited by Singapore’s Urban Planners Suggests Shading Roads with Trees Can Reduce Street-Level Temperatures by up to 2.6 Degrees Celsius.” SpaceDaily. Accessed August 15, 2026. Where the 2.6 °C comes from: the figure is quoted from the Strategies for Cooling Singapore catalog, and the range originates in a Dresden study rather than in Singapore measurements.
Xu, Lei, Ronita Bardhan, Hao Mei, Srilalitha Gopalakrishnan, Xing Zheng, and Thomas Schroepfer. “Harnessing Street Shade to Mitigate Heat Stress: An In-Situ Parallel Investigation under Extreme Heat Conditions in Tropical Singapore.” Science of the Total Environment 958 (2025): 177864.
Meili, Naika, Xing Zheng, Yuya Takane, Ko Nakajima, Kazuki Yamaguchi, Dengkai Chi, Yue Zhu, et al. “Modeling the Effect of Trees on Energy Demand for Indoor Cooling and Dehumidification across Cities and Climates.” Journal of Advances in Modeling Earth Systems 17, no. 3 (2025): e2024MS004590.
Ow, Lai Fern Genevieve, and Eugenie Chan. “Enhancing Outdoor Comfort in Urban Hot Climates: Investigating the Cooling Impact of Two Tropical Trees through Shade Provision and Transpiration.” Arboricultural Journal 46, no. 4 (2024): 272–291.
Tan, Puay Yok, Nyuk Hien Wong, Chun Liang Tan, Steve Kardinal Jusuf, Mei Fen Chang, and Zhi Quan Chiam. “A Method to Partition the Relative Effects of Evaporative Cooling and Shading on Air Temperature within Vegetation Canopy.” Journal of Urban Ecology 4, no. 1 (2018): juy012. Mind the scale: measured on Eugenia oleina saplings 0.8–0.9 m tall in 35-liter pots on a rooftop, not on street trees. The authors say the result “should thus not be taken as a universal phenomenon of urban trees.”
National Environment Agency. “Historical Air Temperature across Singapore (2023).” Data.gov.sg. Accessed August 15, 2026. Analysis by the author, not a citation. 6,893,129 per-minute readings across 17 stations, grouped by timestamp; scripts and full distributions are kept with the project research notes so the figures re-run from the raw file.
Yang, Chen, Mengju Xie, and Thushini Mendis. “Evaluating the Role of Urban Trees on Building Energy Use: A Global Literature Review.” Landscape and Urban Planning 264 (2025): 105475.
Alonzo, Michael, Peter C. Ibsen, and Dexter H. Locke. “Urban Trees and Cooling: A Review of the Recent Literature (2018 to 2024).” Arboriculture & Urban Forestry 51, no. 5 (2025): 420–444.
Haddad, Shamila, Wanni Zhang, Riccardo Paolini, Kai Gao, Muzahim Altheeb, Abdulrahman Al Mogirah, Abdullatif Bin Moammar, et al. “Quantifying the Energy Impact of Heat Mitigation Technologies at the Urban Scale.” Nature Cities 1 (2024): 62–72.
Centre for Climate Research Singapore. “Climate Change Projections for Singapore.” Chap. 5 in Singapore’s Third National Climate Change Study (V3): Stakeholder Report. Meteorological Service Singapore, 2024. Table 5.1 gives the annual average daily mean air temperature over 2080–2099 as the mean of five models, with the model minimum and maximum in brackets: 29.0 (28.5–29.5) under SSP1-2.6, 29.9 (29.3–30.7) under SSP2-4.5, and 31.7 (30.7–32.9) under SSP5-8.5, against 27.9 °C observed at the current nine stations. The 28.5 and 32.9 above are the outer bounds of that set. Note this is the annual mean: MSS’s online visualiser defaults to a seasonal daily maximum, which is a different and much higher figure.

Energy and Emissions

Energy Market Authority. “Energy Transformation.” Singapore Energy Statistics, chap. 2. Accessed August 16, 2026.
World Nuclear Association. “Nuclear Power Reactors.” Information Library. Accessed August 16, 2026.
“Singapore ‘Seriously’ Considering Nuclear Energy.” World Nuclear News, 2025.
Energy Market Authority. “Korea Hydro and Nuclear Power and Energy Market Authority of Singapore to Cooperate on Civil Nuclear Energy Cooperation.” Media release, March 1, 2026.

Satellite Imagery

United States Geological Survey. “Landsat Collection 2 Level-2 Science Products.” Earth Resources Observation and Science Center. Accessed August 16, 2026.

Maju Forest

Liam, Erin. “More Greenery Retained, Fewer Homes Planned for Maju Forest, Gillman Barracks Sites after Public Feedback.” CNA, August 14, 2026.
Koh, Wan Ting. “More than 1,000 Rally at Hong Lim Park over Maju Forest, Gillman Barracks Plans.” CNA, August 16, 2026.
Liam, Erin, and Natalie Ong. “Nature Groups Call for Stronger Environmental Safeguards over Sunset Way, Gillman Barracks Housing Plans.” CNA, July 10, 2026.
Aurecon Singapore. “EIA at Sunset Way: EIA Report.” Rev. 6. Commissioned by the National Parks Board on behalf of the Housing & Development Board, July 1, 2026.

Ecological History

Corlett, Richard T. “The Ecological Transformation of Singapore, 1819–1990.” Journal of Biogeography 19, no. 4 (1992): 411–420.

Forest Cover

World Bank. “Forest Area (sq. km) — Singapore.” World Development Indicators, drawing on the Food and Agriculture Organization’s Global Forest Resources Assessment. Accessed August 17, 2026.

Jurong Railway

“Start of Jurong Railway Link to Federation.” The Straits Times, September 9, 1963, 9.

Tree Species

Tan, Puay Yok, Nyuk Hien Wong, Chun Liang Tan, Steve Kardinal Jusuf, K. Schmiele, and Zhi Quan Chiam. “Transpiration and Cooling Potential of Tropical Urban Trees from Different Native Habitats.” Science of the Total Environment 705 (2020): 135764.
Li, Haiwei, Yongling Zhao, Chenghao Wang, Diana Ürge-Vorsatz, Jan Carmeliet, and Ronita Bardhan. “Cooling Efficacy of Trees across Cities Is Determined by Background Climate, Urban Morphology, and Tree Trait.” Communications Earth & Environment 5, no. 1 (2024): 754.
Green Garden Integrated Services. “Landscaping Strategies to Cool Down Urban Areas.” May 23, 2025.
Staples, George W., and Craig R. Elevitch. “Samanea saman (Rain Tree).” In Traditional Trees of Pacific Islands: Their Culture, Environment, and Use, edited by Craig R. Elevitch. Hōlualoa, HI: Permanent Agriculture Resources, 2006.
Teo, L. M. R., L. Yu, and X. Luo. “Variations in the Temperature Response of Photosynthesis among Nine Common Tree Species Planted in Singapore.” Frontiers in Forests and Global Change 9 (2026): 1738900.
Chen, Letitia. “Singapore’s Planting Palette: What Trees Colour Our Streets and Parks?” The Straits Times, June 15, 2026.
Ihsan, F., and D. Rosleine. “Cooling Effect to Mitigate Urban Heat Island by Pterocarpus indicus, Swietenia macrophylla and Samanea saman in Bandung, West Java Indonesia.” IOP Conference Series: Earth and Environmental Science 528 (2020): 012057.

The live figures on this page come from the data.gov.sg public APIs.