Paper published on urban greenspace and the indoor environment

The  exposome includes urban  greenspace, which may  affect  health via  a  complex set  of pathways, including reducing exposure to particulate matter (PM) and  noise.  We assessed these  pathways using indoor exposure monitoring data  from  the  HEALS study  in four European urban areas  (Edinburgh, UK; Utrecht, Netherlands; Athens  and  Thessaloniki, Greece).

Methods:  We  quantified  three  metrics  of  residential  greenspace  at   50 m  and   100 m  buffers:   Normalised Difference Vegetation Index  (NDVI), annual tree  cover  density, and  surrounding green  land  use.  NDVI values were  generated for both  summer and the season  during which the monitoring took place. Indoor PM2.5 and noise levels  were  measured by Dylos and  Netatmo sensors, respectively, and  subjective noise  annoyance was collected by questionnaire on  an  11-point scale.  We used  random-effects generalised least  squares regression models to assess  associations between greenspace and  indoor PM2.5 and  noise,  and  an ordinal logistic  regression to model the  relationship between greenspace and  road  noise  annoyance.

Results:  We identified a significant inverse relationship between summer NDVI and  indoor PM2.5 (−1.27 μg/m3
per 0.1 unit  increase [95%  CI -2.38 to −0.15]) using a 100 m residential buffer.  Reduced (i.e., < 1.0) odds ratios (OR) of road  noise  annoyance were  associated with  increasing summer (OR = 0.55  [0.31 to 0.98]) and  season- specific  (OR = 0.55  [0.32 to 0.94]) NDVI levels,  and  tree  cover  density (OR = 0.54  [0.31 to 0.93] per  10 per- centage point increase), also at a 100 m buffer.  In contrast to these  findings, we did not  identify any  significant associations between greenspace and  indoor noise  in fully  adjusted models.

Conclusions:  We  identified reduced indoor levels  of  PM2.5  and  noise  annoyance, but  not  overall noise,  with increasing outdoor levels  of certain greenspace indicators. To corroborate our  findings, future research should examine the  effect of enhanced temporal resolution of greenspace metrics during different seasons, characterise the  configuration and  composition of green  areas, and  explore mechanisms through mediation modelling.