Not always an amenity: Green stormwater infrastructure provides highly variable ecosystem services in both regulatory and voluntary contexts

Green stormwater infrastructure (GSI) is advocated for its potential to provide multiple ecosystem services, including stormwater runoff mitigation, wildlife habitat, and aesthetic value. However, the provision of these ecosystem services depends on both facility design and maintenance, which may vary based on whether GSI was installed to fulfill regulatory construction permit requirements or implemented voluntarily as part of urban greening initiatives. We evaluated 76 GSI facilities distributed across Baltimore, MD, USA, comprising 48 voluntary and 28 regulatory facilities. Each facility was scored on indicators related to the provision of stormwater, habitat, and aesthetic ecosystem services, as well as the presence of educational signage. Ecosystem service scores were highly variable, reflecting a wide range of quality and condition, but we found no significant differences between scores for regulatory and voluntary GSI. However, voluntary GSI scores tended to be higher in areas with greater socioeconomic status, while regulatory facilities showed an inverse relationship. Only voluntary facilities included educational signage. Our findings indicate that GSI facilities can degrade quickly, and that official maintenance requirements for regulatory facilities do not guarantee upkeep. Degraded GSI facilities may do more harm than good, becoming both unsightly and ineffective at providing intended stormwater or habitat benefits.

Methods
We identified GSI facilities to evaluate from datasets of regulatory and voluntary GSI installed in Baltimore through 2019. We limited our study to rain gardens, bioswales, and bioretention and micro-bioretention facilities, as these facility types are typically designed to have multiple co-benefits within discrete boundaries and could be compared using the same criteria. We also included similar facilities in the voluntary GSI dataset referred to as stormwater planters and curb bump-outs. There were 101 of these facilities in the regulatory dataset and 185 in the voluntary dataset, some of which were grouped within the same project (e.g., one parking lot might have four micro-bioretention facilities built at the same time). There were 41 regulatory projects and 104 voluntary projects that included the designated facility types. We also limited our study to facilities that could be viewed from public areas, since the aesthetic qualities of these facilities would have community influence. We used aerial imagery and Google StreetView to identify publicly viewable facilities, yielding a set of 28 regulatory projects and 48 voluntary projects with at least one facility that met our criteria.

We visited each of these projects in early autumn of 2019 (8–10 October) to do a visual condition assessment. For projects with multiple GSI facilities that fit our criteria, we scored only the one most visible from our point of access or chose one randomly if they were equivalently visible. We gave each facility scores for multiple variables chosen because of their relationship to stormwater, habitat, and aesthetic values.