Introduction

Urbanization is occurring at an unprecedented rate and scale, and global urban expansion is predicted to increase almost three-fold from 2000 to 2030 (Seto et al. 2010, 2012). Urban areas are characterized by impervious surface such as pavement, concrete, and other human-made structures that dominate the landscape, as opposed to pervious surfaces such as soil or vegetation. Broadly, urban environments appear to reduce species abundance and richness and homogenize the physical landscape and the biological communities that occupy them (McKinney 2006, 2008, Piano et al. 2020). As a result, urbanization may pose a threat to pollinator communities and therefore the critical ecosystem services they provide.
Pollinators are threatened by a number of factors that may relate to or interact with urbanization, including habitat loss, climate change, invasive species, and pesticide use (Winfree et al. 2009, Potts et al. 2010, Vanbergen and Initiative 2013). The increase in impervious surface, spatially disjointed development, and exacerbation of other threats associated with urbanization are hypothesized to have considerable consequences for pollinators and pollination services via destruction and fragmentation of suitable habitat, changes in plant community composition, and loss of nesting resources. However, these potential impacts of urbanization on pollination are understudied relative to other threats, and this has been highlighted as a critical research direction for pollination ecology (Knight et al. 2018). Urbanization has been repeatedly linked to lower pollinator abundance and less diverse pollinator communities (Ahrné et al. 2009, Geslin et al. 2016, Burdine and McCluney 2019, Pereira et al. 2020), with one city in Brazil experiencing a 45% decline in bee species richness over 34 years of urban development (Cardoso and Gonçalves 2018). These relationships, however, are highly variable across species, functional traits, and scales (Bates et al. 2011, Banaszak-Cibicka and Żmihorski 2012, Geslin et al. 2016, Cardoso and Gonçalves 2018).
The ways in which urbanization influences the pollination of urban plants is even less understood (Knight et al. 2018, Wenzel et al. 2020). Urbanization may impact plant-pollinator interactions by changing pollinator abundance, diversity, and behavior (Harrison and Winfree 2015). Changes in the pollinator community may in turn reduce the quantity or quality of pollination services to plants, potentially jeopardizing plant reproduction. For example, pollen limitation occurs when seed or fruit production is limited by inadequate pollen receipt. While pollen limitation is commonly observed in natural populations (Burd 1994, Knight et al. 2005), it appears to be a major driver of reproductive impairment for plants in fragmented habitats, such as those in urban environments (Aguilar et al. 2006). Loss of pollination may be especially pronounced for plants with more specialized floral morphology because their pollination is limited to a narrow subset of pollinators. If those pollinator species decline or are lost, there will be fewer alternative pollinators that can provide pollination services in their absence. For example, across several small conservation areas within an urban matrix in South Africa, the loss of a single bee species resulted in pollen limitation and seed set failure in six orchid species with a specialized pollination system, while a more generalist species was unaffected (Pauw 2007).
Effects of urbanization on plant-pollinator interactions and pollen limitation can have implications for both wild and managed plant communities, such as in urban agriculture. Agriculture worldwide relies heavily on pollinators (Klein et al. 2007), and urban agriculture in particular is becoming increasingly important for urban growth, with almost one fifth of global food production occurring in or near urban areas (Armar-Klemesu 2000). Cities can support equal or even greater pollinator abundance and diversity than intensively used agricultural areas (Hall et al. 2017, Wenzel et al. 2020), making urban agriculture an important prospect in the face of growing food demand and concern over the preservation of plant and pollinator populations. Urban farms therefore provide a relevant and unique framework to study the impacts of urbanization on pollination services.
Here, we aim to assess the effects of urbanization on pollination services across a series of urban farms within the City of Chicago, IL, USA. Specifically, we test whether and how pollinator visitation and pollen limitation differ across varying levels of urbanization, as measured by impervious surface, in two different pollination systems. We focus on two plant species: Solanum lycopersicum (tomatoes; Solanaceae) and Cucurbita pepo (squash; Cucurbitaceae), two agriculturally important crops that each represent a unique pollination morphology and ecology. Our experiment takes place within a network of urban farm sites across a gradient of urbanization, ranging from the urban core of Chicago, Illinois, USA out into the suburbs, to investigate the following questions: (1) Is plant reproduction pollen-limited at urban farm sites? (2) How does urbanization affect the magnitude of pollen limitation? (3) How does urbanization affect pollinator visitation? We expected plants at more urban sites to exhibit the most pollen limitation, and that this effect would be greatest inS. lycopersicum as the more specialized species. We also hypothesized that pollinator visitation to these species would be negatively related to the level of urbanization. Our work helps to fill an important gap in our understanding of how urbanization can impact pollination services to different plant species.