Extracellular Enzyme Patterns Provide New Insights Regarding Nitrogen Transformation Induced by Alkaline Amendment of Acidic Soil

• Published in: Agronomy (Basel) 2022-12, Vol.12 (12), p.3015
• Main Authors: Yin, Junhui, Bai, Xin, Fenton, Owen, Tang, Bingbing, Chen, Shuo, Ma, Yan, Zhang, Shuai, Cao, Wenchao, Ding, Shuai, Liu, Rui, Chen, Qing
• Format: Article
• Language: English
• Subjects: acidic soil; Acidic soils; Acidification; alkaline amendment; Alkaline soils; Aminopeptidase; Ammonia; Ammonification; Availability; Beta cells; Carbon; Enzymatic activity; Enzyme activity; Enzymes; extracellular enzyme activity; Glucosaminidase; Glucosidase; Leucine; Machine learning; Magnesium; Management planning; Mineralization; Nitrification; Nitrogen; pH effects; Potassium; Productivity; Soil acidification; Soil acidity; Soil chemistry; Soil pH; Soils; Statistical analysis; Transformations; Urea; Variance analysis; Xylosidase; α-Glucosidase
• ISSN: 2073-4395; 2073-4395
• DOI: 10.3390/agronomy12123015

Nitrogen (N) availability is generally a limiting factor in highly acidic soil, which could be improved by amending these soils with alkaline materials. Soil extracellular enzyme activity (EEA) plays an important role in N transformation; a current knowledge gap is how this occurs in acidic soils amended with alkaline material. The present 45-day incubation experiment was designed to examine the effects of different amounts of alkaline materials (urea and/or calcium–silicon–magnesium–potassium fertilizer (CSMP)) on N transformation. The results show that soil pH significantly increased after the CSMP amendment (1.2 units) and increased soil net N mineralization (Rmin), net nitrification (Rnit) rates, and net ammonification (Ramo) rates. CSMP amendment changed the different soil EEA but with differing or opposing effects, e.g., Rnit was positively correlated with the activities of L-leucine aminopeptidase, β-xylosidase, α-glucosidase, and N-acetyl-β-glucosaminidase but negatively correlated with β-1, 4-glucosidase and β-cellobiosidase. A machine learning analysis indicated that the best predictor for Rmin and Ramo was soil pH, and for Rnit, it was nitrate. The results of the present study improve our understanding of N availability in acidic soils amended with materials to control soil pH. Such knowledge could lead to more bespoke nutrient management planning at the field scale, leading to better agronomic and environmental outcomes.