Dissimilatory Nitrate Reduction Processes in Typical Chinese Paddy

Using soil slurry-based 15N tracer combined with N2/Ar technique, the potential rates of denitrification, anaerobic ammonium oxidation (anammox), and ...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/est

Dissimilatory Nitrate Reduction Processes in Typical Chinese Paddy Soils: Rates, Relative Contributions, and Influencing Factors Jun Shan,† Xu Zhao,† Rong Sheng,‡ Yongqiu Xia,† Chaopu ti,† Xiaofei Quan,†,§ Shuwei Wang,†,§ Wenxue Wei,‡ and Xiaoyuan Yan*,† †

State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008, China ‡ Key Laboratory of Agro-ecological Processes in Subtropical Regions and Taoyuan Agro-ecosystem Research Station, Soil Molecular Ecology Section, Institute of Subtropical Agriculture, Chinese Academy of Sciences, Changsha 410125, China § University of Chinese Academy of Sciences, Beijing 100049, China S Supporting Information *

ABSTRACT: Using soil slurry-based 15N tracer combined with N2/Ar technique, the potential rates of denitrification, anaerobic ammonium oxidation (anammox), and dissimilatory nitrate reduction to ammonium (DNRA), and their respective contributions to total nitrate reduction were investigated in 11 typical paddy soils across China. The measured rates of denitrification, anammox, and DNRA varied from 2.37 to 8.31 nmol N g−1 h−1, 0.15 to 0.77 nmol N g−1 h−1 and 0.03 to 0.54 nmol N g−1 h−1, respectively. The denitrification and anammox rates were significantly correlated with the soil organic carbon content, nitrate concentration, and the abundance of nosZ genes. The DNRA rates were significantly correlated with the soil C/N, extractable organic carbon (EOC)/NO3− ratio, and sulfate concentration. Denitrification was the dominant pathway (76.75−92.47%), and anammox (4.48−9.23%) and DNRA (0.54−17.63%) also contributed substantially to total nitrate reduction. The N loss or N conservation attributed to anammox and DNRA was 4.06−21.24 and 0.89−15.01 g N m−2 y−1, respectively. This study reports the first simultaneous investigation of the dissimilatory nitrate reduction processes in paddy soils, highlighting that anammox and DNRA play important roles in removing nitrate and should be considered when evaluating N transformation processes in paddy fields.

1. INTRODUCTION Nitrogen (N) is generally the key element limiting rice production; thus, there has been much research on the factors that affect N retention and loss in paddy fields.1,2 The waterlogged conditions during the main stage of rice growth provide a unique environment for dissimilatory nitrate reduction processes including denitrification, anaerobic ammonium oxidation (anammox), and dissimilatory nitrate reduction to ammonium (DNRA).2,3 Among these processes, denitrification has been intensively studied4,5 and was considered to be the major pathway converting fixed N to N2 until the discovery of anammox, in which ammonium is oxidized to dinitrogen using nitrite as an electron acceptor under anaerobic conditions.6−8 Much of what is known on anammox comes from aquatic environments, including marine sediments,9 river estuaries,10−12 freshwater and lake sediments,13,14 where anammox significantly contributes to N loss (up to 79% of the total N2 production).15 Studies on the occurrence and role of anammox in terrestrial ecosystems are limited, although diverse anammox bacteria have been found in agricultural soils.16 Until now, few studies have reported anammox activity © 2016 American Chemical Society

and estimated the importance of anammox in N loss in paddy and farm soils, where anammox accounts for 0.4−37% of the total N 2 production depending on the soil type and profile.17−23 In addition to anammox, DNRA has also been recognized in soil and may act as the dominant pathway for the removal of nitrate under specific conditions (e.g., low redox potential and high ratio of electron donor to electron acceptor).24,25 Only three studies have reported DNRA rates in paddy soils, showing that DNRA contributed 3.9−25.4% of total nitrate consumption, depending on the abundance of DNRA bacteria and the environmental conditions.26−28 It is possible that DNRA is coupled with anammox by providing ammonium.29 By converting nitrate to N2, both denitrification and anammox result in a loss of N, whereas DNRA provides ammonium for Received: Revised: Accepted: Published: 9972

April 9, 2016 June 14, 2016 August 6, 2016 August 6, 2016 DOI: 10.1021/acs.est.6b01765 Environ. Sci. Technol. 2016, 50, 9972−9980

Article

Environmental Science & Technology

tracer technique, (ii) elucidate the critical factors that influence the dissimilatory nitrate reduction processes, and (iii) assess whether the rates of denitrification and anammox from slurrybased 15N tracer experiments can approximate the N removal rates without utilizing 15N-substrate in typical paddy soils in China.

rice uptake and microbial immobilization, leading to N retention in paddy soils. In a single habitat, competition among denitrification, anammox and DNRA is expected; however, previous studies related to dissimilatory nitrate reduction processes in paddy soils focused on one or two independent processes (e.g., denitrification and anammox or DNRA).18,20,26 Simultaneous investigation of the rates of denitrification, anammox, and DNRA, and their relative contributions to total nitrate reduction is lacking. In marine and estuarine ecosystems, the environmental controls governing the competition among different dissimilatory nitrate reduction processes include nitrate concentration, the availability of organic matter, temperature, and the sulfide concentration.24,30 However, the factors regulating denitrification, anammox, and DNRA, and their relative contributions to dissimilatory nitrate reduction process in paddy soils are unknown. Soil microbial communities related to nitrate reduction play a crucial role in controlling rates of denitrification, anammox, and DNRA, because these processes are mainly mediated by microbial activities.31 Microbial nitrate reduction functional genes, such as narG, nirS/nirK, norB, nosZ, and hzsB are commonly used as molecular markers to characterize the denitrifying and anammox bacteria community and to indirectly reflect the activities of dissimilatory reduction processes.32−34 Despite the potential importance of the denitrifying microbial community to the rates and products of the dissimilatory reduction processes, no clear linkages have been found between the community composition or the functional gene abundance and the rates of the dissimilatory reduction processes.31 Previous studies have shown that anammox rates were significantly correlated with the abundance of hzsB genes,20,21 but little is known about the relationships between denitrification or DNRA rates and the microbial nitrate reduction functional genes in paddy soils. Laboratory slurry-based15N tracer techniques are often used as standard methods to determine the potential rates of denitrification, anammox and DNRA following the addition of either 15NO3− alone or in combination with 15NH4+ to the sediments and soils.35−37 In order to achieve high levels of 15N enrichment, the addition of 15NO3− or 15NH4+ is often excessive, which consequently increases the availability of N and may lead to an overestimation of dissimilatory nitrate reduction rates especially in N-limited systems.38 In addition, direct N2 flux measurement can be precisely achieved by membrane inlet mass spectrometry (MIMS) and the N2/Ar technique,39 which is a relatively benign method without the addition of 15N substrates, although it cannot distinguish N2 production from specific processes, such as denitrification and anammox.39 Together with the intact soil/sediment core incubation method, this technique has been successfully applied to quantify the in situ N removal rates of submerged ecosystems (i.e., the denitrification rates therein).41−43 However, no comparison has been made between the laboratory slurry-based 15N tracer technique and the soil core incubation-based N2/Ar technique, and whether the rates of denitrification and anammox from slurry-based 15N tracer experiments can approximate the activities of the N removal rates without 15N-substrate addition is unknown. Hence, the objectives of the present study were to (i) simultaneously determine the potential rates of denitrification, anammox and DNRA, and evaluate their relative contributions to total nitrate reduction using a laboratory slurry-based 15N

2. MATERIALS AND METHODS 2.1. Soil Sampling. Eleven paddy soil samples were ̀ to 127°11N ̀ and 102°45E ̀ to collected across China (21°0N ̀ ) in July 2014 (see details in Table S1 of the 127°11E Supporting Information). These soil samples were chosen because their locations could mainly represent the dominant rice production areas in China. Urea was the routinely chemical N fertilizer in these fields. For sites CS, LY, YX, JD, YA1, and YA2, 240−300 kg urea-N ha−1 were applied, while for site HL, YT, ZJ, JMS, and LA, 120−180 kg urea-N ha−1 were applied in the rice season. All soil samples were collected at 0−20 cm depth from field plots with at least three replicates (ca. 2−4 kg for each subsample) for each location and were immediately transferred to sterile plastic bags with ice packs (4 °C) and were shipped to the laboratory as soon as possible. Each soil sample was divided into three parts: one subsample was air-dried and sieved to