Data for: Spatio-temporal land use dynamics filter life history strategies to shape urban spontaneous plant assemblages

We set up 79 sampling sites across the city following a systematic sampling design (Figure 1), excluding two sites that were inaccessible. The sampling sites were initially set up by Qian et al. (2020) to study the composition of urban spontaneous plant assemblages, and each site was a 1×1 km square grid and 3 km apart from each other, in order to avoid local potential biases caused by spatial autocorrelation of plant distributions.
The field surveys were conducted in two continuous growing seasons: from August to October in 2019, and from March to May in 2020. Within each site, we selected five typical habitat types where spontaneous plants commonly occur: vacant lots, gravel lots, road verges, forest edges, and shrub edges. Depending on the patch size of each habitat type per site, we placed 6 to 43 quadrats with the size of 1×1 m (2×2 m in cases where > 1 m tall plants were present) to record species presence and abundance. During the two sampling seasons, a total of 1460 quadrats were set up and surveyed across our sampling sites.
We studied eleven functional traits that characterize major life history strategies of herbaceous species, including life cycle, seed dispersal ability, seed size, plant maximum height (), chlorophyll content, specific leaf area (SLA), mass-based leaf carbon, nitrogen, and phosphorus concentrations, as well as the C:N and N:P ratios (Table 1). Qian et al. (2020) reported the presence of 279 spontaneous plant species across our sampling sites, yet it was infeasible to measure traits for all present species, especially when there were over 100 species rarely distributed in less than 5% of sites. Thus, we selected 54 species that could be commonly found and had sufficient individuals for trait measurements. Among our focal traits, life cycle, seed dispersal ability, and seed size were retrieved from the Flora of China (www.eFloras.org). Seed dispersal ability was assessed by its dispersal syndrome (e.g., autochory as short dispersal and zoochory as long dispersal). We measured the remaining traits following the standard measurement protocols as described in Pérez-Harguindeguy et al. (2013). Plant height was measured from the ground level to the upper boundary of the main photosynthetic tissues on a plant. Foliar traits were measured on five intact and sun-exposed leaves from one to five mature individuals per species depending on availability. The chlorophyll content was measured in situ by a SPAD-502 Plus chlorophyll meter (Konica Minolta Inc, Japan). All leaf samples of each species were then pooled and sent to the laboratory, where leaf areas were immediately scanned and measured with the ImageJ software (ver. 1.51j8; NIH, USA) upon arrival. Next, leaf samples were oven-dried at 80℃ for 48 h and ground to measure the leaf stoichiometry and dry mass. The SLA was calculated as the area of fresh leaf divided by its dry mass. Leaf carbon and nitrogen concentrations were measured with a Shimadzu TOC-TN analyzer (Shimadzu Scientific, Japan; acid digestion pre-processing for N analysis), and phosphorus concentration was measured by the molybdenum anti-colorimetric method after acid digestion, see Hu et al. (2021) for detailed procedures in trait measurement. Trait measurements across sites were then aggregated into the species-level mean trait values as we primarily focused on inter-specific trait variation here.