Effect of pollution and climate factors on correlation between readings from low-cost sensors and AURN background station. Graphs show variation in Pearson coefficient between the three sensor models and the background AURN reference station per site with (1) months (excluding the Alphasense OPC-N2 for AQM B.2 due to the lack of data), (2) quartile of background PM2.5 concentration (μg/m3), (3) quartile of relative humidity (RH), (4) wind direction and (5) quartile of temperature. August has been excluded from the graphs except for (1). Each box represents median, the 25th and 75th percentiles with maximum and minimum whiskers of Pearson coefficient for the locations considered (n = 4 for the Alphasense OPC-N2 and n = 6 the Plantowers). Data analysed using a Friedman analysis of variance (ANOVA) with Dunn’s post-hoc test for pairwise comparison, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.