Authored by Dr Jo Garrett

One of the biggest conundrums related to nature and health rests on an important question: how do the environments where people live affect their wellbeing?

There are lots of ways to start answering that question, and for researchers who specialise in data and geography a common approach has been to use something called a Lower-layer Super Output Area (LSOA).

It’s a bit of a mouthful, but essentially LSOAs allow us to divide the country up into smaller areas that have roughly the same number of people in them. This makes it much easier (and more accurate) to compare statistics between different parts of the country, like measuring deprivation for example.

But there’s a catch, LSOAs aren’t all the same size. Because their boundaries are based on population, not land area, urban LSOAs are usually much smaller than their rural equivalents: people tend to live in much denser concentrations – and therefore smaller spaces – in cities.

Researchers often take environmental features, like the amount of green space, and calculate an average for the whole LSOA. Then they assign that average to everyone living there. The problem? Conditions can vary a lot within an LSOA, especially in rural areas where boundaries cover large areas.

So, we’ve devised a new way to more realistically assess the environmental features where people live.

Using the Living England Habitat Map as an example, we combined LSOA and postcode-level data to account for varying area sizes (as well as other, somewhat technical issues). We then compared our method with the typical approach, which calculates an average at the small geography unit level.

Overall, our new method suggested that people were actually exposed to less green and natural land cover compared to the typical LSOA averaging approach. Whilst their exposure to built-up areas was higher by a staggering 10% on average.

However, these patterns also varied across different parts of the country, and according to land cover type, and LSOA size.

We’re really excited by the possibilities of this new approach and think it could offer a more consistent way to estimate neighbourhood exposure to nature.

We’ve provided all the code and resulting land cover maps so that other researchers can use it, and the paper is available to read open access in the International Journal of Health Geographics at doi.org/10.1186/s12942-025-00425-7.

This blog post was supported by AI tools to improve clarity, readability, and structure. AI suggested changes were all reviewed and edited where necessary to ensure accuracy. All ideas and interpretations are those of the author.