Synthesis:

(Assemblage Distribution)

The preferential distribution of the SNH assemblages according to total metal loads results in a directional (beginning with areas of high soil metal load) pattern of succession on the heterogeneous soils associated with historic fill. In addition, the areas of lower soil metal load appear to facilitate herbaceous wetland assemblages (SEM) that are not dominated by the invasive P. australis. Continued observation of the SEM wetlands is needed to determine if these wetlands are stable. These two patterns provide some support for our second hypothesis by demonstrating that soil metal load correlates with assemblage distribution. However, that correlation was only significant for two of the ten identified assemblages and, in fact, the distributions of the CRM, CRW, and MS assemblages, dominated by aggressive metal tolerant species, were indifferent to total soil metal load.

(Assemblage Trajectories)

Historic aerial photographs were digitized and the four guilds, hardwood forest, wetlands, shrub and herbaceous plants, were mapped using ArcMap (ArcGIS, 2004). An examination of the results indicates that assemblage trajectories in areas where the TML was high (3 or above) favored rapid development of B. populifolia or Populus sp. dominated assemblages. The rapid development of the hardwood assemblage in these areas appears to inhibit the development of shrubs at the higher TML. Interestingly in both scenarios the hardwood community overtakes the herbaceous communities in the 28th year. In addition, guild trajectories between year 31 and 34 are relatively flat. This could be an early indication that the current assemblages are entering a period of stasis. Hence, models for assembly rules, at least those associated with the degraded environments of the urban context, must account for abiotic filters rather than focusing primarily on competition or facilitation between species.

(Metal Attenuation )

Since it is known that metals are often adsorbed or occluded by carbonates, organic matter, Fe-Mn oxides and primary or secondary minerals (Adriano, 1986; Ross, 1994), and since there have been no human impacts at the site that would change the concentration of Fe or Mn, we expected that the change in soil metal concentrations at the site was due primarily to interactions of the historic fill with the developing plant assemblages. It has been demonstrated that leaf litter can provide a sink for metals, which bind passively to organic surfaces or actively through the physiological activity of the microbial colonizers (Gadd, 1993; Ledin, 2000). Leaf litter can also act as a source when microbial activity mobilizes the metal (Gadd, 1993) or through the action of deposit feeders (Weis and Weis 2004). Hence, metal sequestration via plants is dependent upon the rates of uptake and retention by the various tissue types, translocation to deposit feeders and release through decomposition. In this study in general, the concentration of metals has decreased substantially over ten years. Since the potential for offsite transport is minimal and translocation has been demonstrated, it appears that vegetative assemblages are playing a significant role through the sequestration of metals, thereby enhancing the rate of attenuation.

So what?