*3.1. Inorganic Soil Nitrogen Production*

For all landscape patches, median whole core net mineralization and nitrification were highest in the Agr soils, with median rates of 1.1 mg and 0.7 mg N kg/d, respectively (Figure 1; Table S2, Supplementary Materials). Whole core net mineralization and nitrification were next highest in the PP soils (median 0.8 and 0.5 mg N kg/d, respectively). The relatively high inorganic N production rate results for the PP patch could be related to the ability of perennial peanut to fix atmospheric N in the soil in the presence of rhizobia bacteria [25]. In the process of N fixation, atmospheric dinitrogen is cleaved and enzymatically catalyzed to produce two molecules of ammonia, leading to subsidies of NH4 <sup>+</sup> to the soil. When soil NH4 <sup>+</sup> levels are initially low, net mineralization is slow because inorganic N is preferentially incorporated into microbial biomass [26]. On the other hand, increasing levels of soil NH4 <sup>+</sup> have been shown to increase N mineralization rates in soils [26]. In this way, it is likely that N fixation by the perennial peanut led to high initial levels of soil NH4 +, which in turn promoted the high N mineralization rates seen in the PP soil. We are aware of no studies on the fate of fixed N when perennial peanut is used as a turfgrass alternative in urban landscapes. However, as discussed below, exchangeable NO3 − was lower in the PP soils than all other landscape patches, indicating that even though the PP soils had high levels of NO3 − production (nitrification), they were not maintaining high NO3 − levels. Potential fates of the produced NO3 − to explain its loss in the PP soils include denitrification, plant uptake, and leaching. The well-oxygenated and low organic matter conditions of our sandy soils make denitrification unlikely, since denitrification requires oxygen-limited environments and an organic carbon source [27]. We recommend future studies to investigate whether the inorganic N produced in soils under perennial peanut is readily lost via leaching, especially in sandy soils where NO3 − mobility would be high. While perennial peanut offers water-saving advantages over traditional turfgrass [28], it may come with the tradeoff of greater inorganic N leaching potential, though this has not been studied yet for urban soils.

**Figure 1.** Net N mineralization (**a**) and nitrification (**b**) rates in landscape patches. Values based on whole-core basis. Agr; remnant agricultural field, TfGr; turfgrass, PP; perennial peanut, OMx; patches with a mixture of ornamental grasses, OGa; patch with gamma grass, MulP; patch with pine needle mulch, MulC; patch with pine bark nuggets.

After Agr and PP, the TfGr soils had the third highest median rates of net N mineralization and nitrification (Figure 1; Table S2). There were likely fertilizer subsidies of inorganic N in both the Agr and TfGr patches. As with the PP soils, relatively high initial levels of inorganic N for Agr and TfGr probably led to higher inorganic N production rates during the soil incubations when compared to non-perennial peanut turfgrass alternatives (OMx, Oga, MulP, and MulC patches) [26]. For all landscape patches except the mulched areas (MulP and MulC), regression analysis revealed a significant linear relationship between nitrification and net mineralization on a whole-core basis (Figure 2). This is not surprising since nitrifying bacteria need ammonia as an energy source thus that nitrification generally increases with increasing ammonia levels in soils [27]. Identifying the underlying mechanism responsible for the lack of this relationship under the mulched landscape areas was beyond the scope of this paper, but factors that may inhibit nitrification include soil C:N ratios above ~22, acid soil pH conditions, and low oxygen availability [15]. We observed that the MulP landscape patch had the highest C:N among all landscape types (~22–23, Table S1), possibly indicating suppressed nitrification and a relatively less active N cycling microbial community in the MulP soils.

**Figure 2.** Regressions of nitrification against net N mineralization in the landscape patches, on a whole-core basis. Agr; remnant agricultural field, TfGr; turfgrass, PP; perennial peanut, OMx; patch with a mixture of ornamental grasses, OGa; patch with gamma grass, MulP; patch with pine needle mulch, MulC; patch with pine bark nuggets.

#### *3.2. Exchangeable Inorganic Soil Nitrogen*

Among the non-agricultural soils (all patches except Agr), the PP soils contained the highest median KCl-exchangeable NH4 <sup>+</sup> (1.4 mg/kg) but the lowest KCl-exchangeable NO3 − (0.4 mg/kg) on a whole-core basis (Figure 3). As discussed above, this relatively low exchangeable NO3 − pool for the PP soils indicates that N losses via leaching may be taking place. While the use of perennial peanut for urban landscapes has not been studied in terms of its potential for increased N leaching, several studies have looked at this for agricultural settings where perennial peanut is used for forage crops or as a cover crop. Woodard et al. [29] observed that a forage system with perennial peanut leached more N than a comparable one with bermudagrass. In that study, the authors concluded that perennial peanut might pose threats to groundwater quality under agricultural systems because of its apparent role in subsidizing high NO3 − levels in the soil.

**Figure 3.** Exchangeable soil ammonium (**a**) and nitrate (**b**) in the landscape patches. Values are on a whole-core basis. Agr; remnant agricultural field, TfGr; turfgrass, PP; perennial peanut, OMx; patch with a mixture of ornamental grasses, OGa; patch with gamma grass, MulP; patches with pine needle mulch, MulC; patch with pine bark nuggets.

The mulched landscape patches (MulP and MulC), for which vegetation was absent, contained higher mean pools of exchangeable NH4 <sup>+</sup> and NO3 − than the turfgrass and ornamental patches, though there was considerable spread in the data (Figure 3). It is likely that the mulched landscapes had higher mean exchangeable inorganic N due to the lack of a plant sink to remove inorganic N from the soils [18]. It is also likely that the mulches contained their own pools of N, which leached from the mulch and added to soil N pools [30]. In a study of several urban landscape alternatives, Loper et al. [30] observed that organic soil amendments such as composts could increase inorganic N leaching when compared to soils without amendments. In that study, the authors recommended that landowners did not apply nutrients to mulched areas and that landscape managers recognize that different patches of the same landscape may have different management needs.

In general, exchangeable NH4 <sup>+</sup> was highest in the surface 0–15 cm for all soils, while NO3 − was generally higher with soil depth, perhaps indicating a tendency for NO3 − to move downward in the soil profiles that are predominantly sand (Figure 4). At the 30–52 cm soil depth, exchangeable NO3 − was higher in 4 out of 6 non-agricultural soils than it was in the Agr soil, indicating that the urban landscape patches may pose a greater threat of NO3 − leaching than the agricultural soil (Figure 4). Turfgrass N fertilization has been found to contribute to the leaching of NO3 − in Florida's sandy soils. For example, Shaddox and Surtain [31] found that applied N in Bermuda grass turf leached between 8% and 12% on a 10% slope. Notably, the NO3 − concentrations in the 30–52 cm depth for the mixed ornamental (OMx) and the gamma grass patch (OGa) were in sharp contrast to each other, with much more NO3 − at depth for the mixed ornamental patch than for the gamma grass patch. Gamma grass is one of the most deeply rooted ornamental grasses, with root depth up to 1.8 m [32], and its deep rooting may be contributing to N uptake deep in the soil profile.

**Figure 4.** Exchangeable soil ammonium (**a**) and nitrate (**b**) in each landscape patch by soil depth. Agr; remnant agricultural field, TfGr; turfgrass, PP; perennial peanut, OMx; patch with a mixture of ornamental grasses, OGa; patches with gamma grass, MulP; patch with pine needle mulch, MulC; patch with pine bark nuggets.

#### **4. Management Implications and Future Research Needs**

It is beyond the scope of this paper to make recommendations about which landscape patch is preferable from a N availability or groundwater protection standpoint, as our main intent was to only investigate fine-scale variability in N production in a mixedvegetation urban landscape. However, some key takeaways are manifest in our results. The first of these is that inorganic N availability and production were, in fact, highly variable among landscape patches at scales of a few meters, in support of our hypothesis. This finding supports the habitat structure framework of urban soil ecology presented by Bryne [19], who argued that the high spatial heterogeneity of human activities in urban landscapes leads to a large range of soil properties at often small scales. The presence of soil amendments such as mulches or plants such as perennial peanuts that can subsidize soil N pools is just two ways that humans influence urban soil properties. We have shown here that inorganic N production is variable in a single urban landscape and have presented some discussion as to why this may be the case for our specific landscape patch types. At a minimum, we argue that urban landscapes should not be broadly managed but that landscape managers should recognize that the nutrient needs of patches within a single landscape may be different and that the nutrient fates in soil under those patches may also be different. This has important implications for how urban soil properties are included in models that aim to predict nutrient transport from urban landscapes. Very likely, we need spatially finer-scale investigations of nutrient fate and transport in urban landscapes, especially those with mixed vegetation or heterogeneous human influences. Too often, urban landscapes are assumed to be covered with turfgrass only, and other possible land covers need to be more fully considered by those who aim to quantify the effects of urbanization on the environment.

A second key takeaway of this work is the implication related to potential N losses by leaching from certain landscape patches. The high nitrification potential of perennial peanut seen here suggests that this turfgrass alternative may produce NO3 − that can be readily leached from sandy soils. However, our sample size was small, and similar work should be conducted in the future to expand the temporal and spatial scope of this work. In particular, we recommend studying N leaching under perennial peanut when it is used as an urban groundcover. We also recommend future work to better understand the mechanisms that drive the differences in potential N losses we observed among landscape patches. Some of the potential drivers include irrigation status, soil pH, moisture content, organic matter content, and variations in the soil microbial community. Finally, because urban landscapes are tied to human activities, we recognize that sociocultural variables may drive differences in soil properties as well [33,34]. These sociocultural variables may include the choice of plants that are used in ornamental beds. For example, we observed

that the two ornamental patches had very different levels of inorganic N at the greatest soil depth, likely due to greater uptake of N at depth by the deeply rooted gamma grass. This human choice as to which plants to include in a landscape may therefore impact N losses by leaching through the soil profile.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/nitrogen3010009/s1, Table S1, Basic soil physical/chemical properties for landscape patches used in this study; Table S2, Raw data showing nitrification, mineralization, and exchangeable inorganic N for all landscape patches and soil depths.

**Author Contributions:** Conceptualization, M.G.L. and J.B.; methodology, M.G.L.; formal analysis, J.B. and M.G.L.; investigation, J.B.; resources, M.G.L.; data curation, J.B.; writing—original draft preparation, J.B.; writing—review and editing, M.G.L.; visualization, J.B.; supervision, M.G.L.; project administration, M.G.L.; funding acquisition, M.G.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded in part by the Center for Land Use Efficiency at the University of Florida.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data used in this manuscript is available in Table S2.

**Conflicts of Interest:** The authors declare no conflict of interest.
