Evaluating the diffusive gradients in thin-films technique for measurement of plant available nitrogen in soils

Abstract

Accurate measurement of nitrogen (N) as nitrate and as ammonium in soils is crucial for effective agricultural management, as it allows farmers to optimize fertilizer application and prevent nutrient imbalances, thereby maximizing crop productivity while minimizing environmental impact. Here we report the findings from a series of laboratory and greenhouse experiments that aimed to evaluate the ability of the DGT technique to measure N as nitrate (NO3-DGT) and as ammonium (NH4-DGT) in soils, using binding layers containing A520E anion exchange resin and Microlite® PrCH cation exchange resin, respectively. The DGTs were deployed in a series of laboratory experiments using ten soils ranging in chemical and physical properties, amended with different concentrations of NO3-N and NH4-N (equivalent to 75 – 450 kg N/ha). The DGT technique was compared to the more commonly used 2 M KCl extraction for measuring these N species. A subsequent greenhouse experiment sought to examine the ability of NO3-DGT and NH4-DGT to predict yield and N uptake by wheat. Here, four contrasting soil types, each amended with four different rates of N (0 – 375 kg N/ha) were used to examine the response of the wheat plants to different N levels in the soils. Soil concentrations of NO3-N and NH4- N were measured using DGT and 2 M KCl extractions at sowing, and the results were compared against the relative yield and N uptake by the wheat plants at the late flowering stage. The NO3-DGT and NH4-DGT achieved superior detection limits for NO3-N (6.90 µg/L) and NH4-N (6.23 µg/L) when compared to the KCl extraction in unfertilised soils. The N accumulation by the resin gel over 24 h deployments was linear across the N amended soils. The concentrations measured by DGT, and KCl extraction correlated with each other for both species, but the correlation for NO3-N (R² = 0.53; p < 0.001; n = 20) was stronger than for NH4-N (R² = 0.20, p = 0.045; n = 20). The concentration of soil mineral N (i.e. NH4-N + NO3-N) measured by DGT and the KCl extraction predicted around half of the variability in relative yield (both R² = 0.50; p < 0.010; n = 64); however, concentrations of individual mineral N species (NO3-N and NH4-N), were a much weaker predictor of yield and in some cases, a significant relationship could not be identified (R² = 0.01 - 0.43). The uptake of N by wheat could be predicted by both KCl extraction and DGT-measured mineral N, and NO3-N (R² = 0.63 - 0.82; p < 0.01; n = 64). The strongest relationships between soil mineral N and NO3-N concentrations and wheat N uptake were found in the KCl extracts (R² > 0.81, p < 0.001; n = 64). The A520E and PrCH binding layers can provide robust measurements of soil mineral N species across a range of soils and N concentrations. However, some of the challenges posed by changes in N species during soil sample preparation and subsequent deployment (e.g., mineralization of organic N and nitrification) need to be overcome before DGT can be used for predicting N uptake by wheat.