Tranquility / Urban Heat Index: Technical Whitepaper
Full statistical method, bin performance analysis, temporal consistency testing, and regional results across 272,958 property sales.
By Luke Metcalfe, Microburbs Research. February 2026
1. Abstract
This paper presents a composite index measuring urban heat island effects in Australian suburbs. The index combines three density-related variables including measures of vertical development and housing type into a single score. Suburbs with lower density and more detached housing score higher. Suburbs dominated by high-rise towers and apartment blocks score lower.
We trained a statistical model on historical property sales data to predict 4-year annualised growth rates. The model achieved a full R-squared of 0.118 with an out-of-sample test R-squared of 0.0647. The top-scoring suburbs outperformed the national median by +3.1 percentage points over 4 years. The bottom-scoring suburbs underperformed by -2.9 percentage points.
The signal was tested across 163 quarterly periods, 24 individual sample dates, and 13 GCCSA regions. It held in 149 of 163 quarters (91%). It was statistically significant at 22 of 24 sample dates. And it produced a positive spread in 8 of 13 regions.
These results point to a persistent relationship between housing density patterns and subsequent property growth. Lower-density suburbs grow faster. Consistently.
2. Methodology
2.1 Feature Construction
The Tranquility / Urban Heat Index is built from three density-related variables including measures of vertical development and housing type at the suburb level. We do not disclose the specific field names. The combination itself represents proprietary research.
Each variable was standardised and combined into a single composite score using a weighted formula. The weights were derived from a predictive model trained to predict 4-year annualised property growth relative to the national median.
2.2 Model Training
The model was trained on property sales data spanning 2008 to 2021. Each observation is a single property sale. The target variable is the annualised 4-year growth rate from the sale date. The features are the three density-derived variables for the suburb where the sale occurred.
The model was trained using a strict out-of-sample framework. Training and test sets were split by time period to prevent data leakage. Properties sold in overlapping windows were excluded from the test set to ensure clean separation.
2.3 Performance Metric
The primary metric is the difference in median annualised 4-year growth between each bin and the national median. Statistical significance is assessed using a two-sided t-test against the null hypothesis that the bin's mean growth equals the national mean.
t-statistic = (mean(bin) - mean(national)) / SE(bin)
p-value from two-sided t-test
2.4 Out-of-Sample R-squared
The model achieved an out-of-sample test R-squared of 0.0647 and a training R-squared of 0.0929. The full model R-squared is 0.118. This means roughly 6.5% of variance in 4-year suburb-level growth can be explained by three density variables alone on unseen data. Section 7 discusses why this level of explanatory power is still useful for investment decisions.
Note on R-squared
Property prices are influenced by hundreds of factors. Interest rate cycles, infrastructure investment, zoning changes, population growth, local employment, and macroeconomic conditions all play a role. No single thematic index will produce a high R-squared. The relevant question is whether the signal is statistically significant and consistent, not whether it explains most of the variance. With only three input variables, an R-squared of 0.065 out of sample is a strong result.
3. Bin Performance
The model sorts suburbs into three bins based on their Tranquility / Urban Heat score. Each bin has a distinct growth profile.
| Bin | Score Range | Diff vs National | p-value | N (Sales) | Significant |
|---|---|---|---|---|---|
| Top | 81 to 100 | +3.1% | 1.7e-122 | 54,228 | Yes |
| Middle | 17 to 81 | -0.4% | 4.4e-03 | 173,422 | Yes |
| Bottom | 0 to 17 | -2.9% | 7.3e-64 | 45,308 | Yes |
Key observation
All three bins produce statistically significant results. The spread between top and bottom is 6.0 percentage points over 4 years. The monotonic ordering (top positive, middle near zero, bottom negative) confirms the index captures a real gradient in growth outcomes. The top bin's p-value of 1.7e-122 is among the strongest of any Microburbs index.
4. Temporal Analysis
A signal that works at one point in time could be a fluke. We tested the Tranquility / Urban Heat Index across every quarter from 2008-Q1 to 2021-Q3.
Quarterly consistency
The top quartile sits above the bottom quartile in 149 of 163 quarters (91%). The separation is strongest during 2013 to 2016, where the top quartile consistently outperforms by over 3 percentage points. Even during the COVID-era boom of 2020-2021, the gap remains positive. This is a signal that holds across market cycles.
4.1 Date-by-Date Consistency
We tested the top bin's outperformance at 24 individual sample dates between 2011 and 2021. At each date, the model was evaluated independently. The result was statistically significant at 22 of 24 dates.
| Sample Date | Outperformance (4yr) | Significance |
|---|---|---|
| 2011-12 | +2.0% | Significant |
| 2012-04 | +2.7% | Significant |
| 2012-10 | +3.5% | Significant |
| 2012-11 | +3.6% | Significant |
| 2013-06 | +4.6% | Significant |
| 2014-02 | +5.6% | Significant |
| 2014-07 | +5.9% | Significant |
| 2015-04 | +4.8% | Significant |
| 2015-06 | +4.4% | Significant |
| 2015-09 | +3.7% | Significant |
| 2016-05 | +3.3% | Significant |
| 2016-08 | +3.1% | Significant |
| 2017-04 | +2.8% | Significant |
| 2017-09 | +2.9% | Significant |
| 2018-01 | +2.8% | Significant |
| 2019-04 | +1.7% | Significant |
| 2019-05 | +1.6% | Significant |
| 2019-06 | +1.5% | Significant |
| 2019-11 | +1.1% | Significant |
| 2019-12 | +0.9% | Not Significant |
| 2020-01 | +0.7% | Not Significant |
| 2021-01 | +3.0% | Significant |
| 2021-03 | +3.2% | Significant |
| 2021-08 | +3.2% | Significant |
Pattern in non-significant dates
The two non-significant results cluster in late 2019 and early 2020. The outperformance was still positive (+0.9% and +0.7%) but the sample sizes at those dates were not large enough to clear the significance threshold. The signal strengthened again in 2021, returning to +3.0% and above. This brief soft patch coincided with a period of macro uncertainty before COVID lockdowns took hold.
5. Regional Robustness
A signal that works only in one city is less useful than one that works nationally. We tested the Tranquility / Urban Heat Index across all 13 GCCSA regions in Australia.
| Region (GCCSA) | Top Q Growth | Bottom Q Growth | Spread | N (Sales) |
|---|---|---|---|---|
| Greater Darwin | -1.23% | -5.11% | +3.88% | 294 |
| Greater Melbourne | +1.29% | -2.21% | +3.50% | 5,440 |
| Rest of Vic. | +2.04% | -0.90% | +2.94% | 7,443 |
| Rest of NSW | +2.65% | +0.15% | +2.50% | 8,656 |
| Greater Brisbane | +2.12% | +0.58% | +1.54% | 6,416 |
| Greater Adelaide | +0.49% | -0.91% | +1.40% | 3,426 |
| Greater Perth | -2.45% | -3.49% | +1.04% | 3,948 |
| Greater Sydney | -0.13% | -1.07% | +0.94% | 5,286 |
| Rest of NT | -5.20% | -5.67% | +0.47% | 326 |
| Rest of Qld | +0.17% | -0.29% | +0.46% | 13,919 |
| ACT | +0.23% | +0.20% | +0.03% | 1,392 |
| Rest of WA | -2.50% | -2.36% | -0.14% | 5,721 |
| Rest of SA | -0.20% | +0.03% | -0.23% | 3,582 |
Regional pattern
The signal produces a positive spread in 8 of 13 regions. Greater Darwin leads with a +3.88% spread. Greater Melbourne follows at +3.50%. Five regions show a negative or negligible spread. The index works best in capital cities with meaningful variation in housing density. It is weakest in regional areas and smaller capitals where housing stock is more uniform.
Greater Sydney is an interesting case
The spread is +0.94%, which is positive but much weaker than Melbourne (+3.50%). Sydney's property market is heavily influenced by land scarcity and transport infrastructure. A suburb's proximity to the CBD and major rail lines drives prices more than housing density alone. In Melbourne, the gap between inner-city apartment precincts and outer detached-house suburbs is wider and more consistent. Sydney's density gradient is compressed by geography. The harbour, coastline, and rail corridors create pockets of high value regardless of housing type. The signal still works in Sydney. But it works harder in Melbourne, where density tells you more about a suburb's growth trajectory.
6. Suburb-Level Evidence
Suburb-level comparisons for selected cities are available on the summary.
7. Defence of Method
7.1 Why a Low R-squared Still Matters
An R-squared of 0.065 (out of sample) means the model explains 6.5% of variance in 4-year suburb-level growth. This is low in absolute terms. It would be a poor result for a model trying to predict exact growth rates. But that is not how this index is intended to be used.
Property prices are shaped by hundreds of factors. Interest rate cycles, infrastructure investment, zoning changes, population growth, local employment, and macroeconomic conditions all play a role. No single thematic index will capture the majority of variance. A model that explained 50% of property growth from three density variables would be implausible and almost certainly overfitted.
The relevant questions are different. Does the signal exist? Is it statistically significant? Does it persist across time and geography? The evidence says yes on all three counts.
7.2 Statistical Significance
The top bin's outperformance has a p-value of 1.7e-122. This is not borderline. The probability of observing a +3.1% difference across 54,228 sales by random chance is effectively zero. For context, a p-value below 0.05 is the standard threshold for statistical significance. The Tranquility / Urban Heat Index exceeds this by 120 orders of magnitude.
7.3 Consistency Over Time
The signal was significant at 22 of 24 sample dates spanning a decade. The two non-significant dates fall in late 2019 and early 2020, during a period of market uncertainty before COVID. A signal that works 92% of the time across a full market cycle is strong. And the 91% quarterly consistency rate (149 of 163 quarters) confirms this is not a seasonal or cyclical artefact.
7.4 Geographic Breadth
The spread is positive in 8 of 13 GCCSA regions. It works in rising markets (Rest of NSW, Greater Brisbane) and falling markets (Greater Perth, Greater Darwin). The five regions with negligible or negative spreads are either very small (Rest of NT, ACT) or have uniform housing stock (Rest of SA, Rest of WA). The signal requires meaningful density variation within a region to produce separation.
7.5 Practical Use
Investors do not need a model to predict exact prices. They need a signal to tilt the odds in their favour across a portfolio of purchases. A 3.1 percentage point advantage per purchase, compounded over time, is substantial. Combined with other Microburbs signals, the Tranquility / Urban Heat Index forms one layer in a multi-factor approach to suburb selection.
Analogy
A weather model that explains 6.5% of daily temperature variation would be useless for predicting tomorrow's temperature. But a model that reliably identifies which regions are hotter on average, across 13 years of data and 13 geographic zones, is describing a real climate pattern. The Tranquility / Urban Heat Index describes a property 'climate' pattern. Low-density suburbs grow faster. That pattern has held for over a decade.
8. Limitations
8.1 Census Data Is Point-in-Time
The index relies on Australian Census data from 2016 and 2021. Census data is collected every five years. Housing density and building types can shift between census dates. A suburb that scored highly in 2016 may have seen significant apartment development by 2021. The model cannot capture these shifts in real time.
8.2 Backward-Looking Model
The model was trained on historical data from 2008 to 2021. Past patterns do not guarantee future results. The relationship between housing density and property growth could weaken or reverse if underlying market dynamics change. Government policy shifts toward densification, for example, could alter the pattern.
8.3 Individual Suburb Variation
Even within the top bin, individual suburb outcomes vary widely. Denham Court scored 100.0 and returned +1.60% per year during 2021-2025. Brunswick East scored 98.3 and returned -2.34% per year in the same period. The index provides a statistical edge across large numbers of purchases, not a guarantee for any single suburb.
8.4 Regional Limitations
The signal does not work in every region. Five of 13 GCCSA regions produced negligible or negative spreads. Rest of SA, Rest of WA, ACT, Rest of NT, and Rest of Queensland all had spreads under +0.50%. Investors in these markets should weight this index lower in their decision-making.
8.5 Sample Size Variability
Some regions have small sample sizes. Greater Darwin contributed only 294 sales. Rest of NT contributed 326 sales. Results in small-sample regions carry wider confidence intervals and should be interpreted with caution.
8.6 No Causal Claim
This paper documents a correlation, not a causal mechanism. We hypothesise that lower-density suburbs attract owner-occupiers with longer holding periods, creating more stable demand. Detached houses on larger lots also appreciate faster because the land component is higher relative to the building. But the data does not prove causation. Other unmeasured variables may explain part or all of the observed relationship.
Summary of limitations
The Tranquility / Urban Heat Index is a statistical tool, not a crystal ball. It identifies a persistent pattern across 272,958 sales, 13 years, and 13 regions. But individual outcomes will vary. Census data updates infrequently. The model is backward-looking. It works in 8 of 13 regions. Use this index as one factor in a broader investment framework.
Access Suburb-Level Scores
Get Tranquility / Urban Heat scores for every suburb in Australia. Combine with other Microburbs signals to build a shortlist backed by data.