Nutrient and toxic elements in soils and plants across 10 urban community gardens: Comparing pXRF and ICP-based soil measurements

Abstract

Urban community gardens are becoming increasingly important to rehabilitate developed lands and combat the lack of access to fresh produce. Portable X-ray fluorescence (pXRF) offers a rapid, cost-effective method for assessing the elemental composition of soils but needs further study to determine its efficacy in urban agriculture. The objectives of this study were to evaluate if pXRF measurements of macronutrients (Ca, K, P), micronutrients (Cu, Mn, Zn), and toxic elements (As, Pb) generate results comparable with traditional soil analyses and if the soil measurements correlate with plant tissue concentrations at 10 community gardens across the eastern United States. From field-condition analyses of soils by pXRF and pseudototal digestions, we observed that both methods provide agreeable estimates of concentrations for some elements (Mn, Cu, Zn, Pb) but not for macronutrients (Ca, K, P). We hypothesize that low accuracy in pXRF measurements and macronutrients within silicates caused the poor agreement between the methods. Sieved and dried soil pXRF concentrations were in strong agreement with field-condition pXRF concentrations, suggesting rock removal and drying did not improve measurements. Our results highlight that pXRF can be an accurate and effective tool for screening for Mn, Cu, Zn, and Pb. Some elements, such as Pb in fruits; Mn, Cu, and Zn in leaves; and Zn and Pb in roots, could be estimated by soil pXRF or inductively coupled plasma-based analyses. Macronutrients were poorly estimated for fruits, leaves, and roots. Instead of soil concentrations, identifying genus-specific and garden-specific factors may be important for generating plant uptake predictive models.