*Article* **How Tillage and Fertilization Influence Soil N2O Emissions after Forestland Conversion to Cropland**

**Xiao Ren 1,2, Bo Zhu 1,\*, Hamidou Bah 1,2 and Syed Turab Raza 1,2**


**\*** Correspondence: bzhu@imde.ac.cn

Received: 22 August 2020; Accepted: 22 September 2020; Published: 25 September 2020

**Abstract:** Soil nitrous oxide (N2O) emissions are influenced by land use adjustment and management practices. To meet the increasing socioeconomic development and sustainable demands for food supply, forestland conversion to cropland occurs around the world. However, the effects of forestland conversion to cropland as well as of tillage and fertilization practices on soil N2O emissions are still not well understood, especially in subtropical regions. Therefore, field experiments were carried out to continuously monitor soil N2O emissions after the conversion of forestland to cropland in a subtropical region in Southwest China. One forestland site and four cropland sites were selected: forestland (CK), short-term croplands (tillage with and without fertilization, NC-TF and NC-T), and long-term croplands (tillage with and without fertilization, LC-TF and LC-T). The annual cumulative N2O flux was 0.21 kg N ha−<sup>1</sup> yr−<sup>1</sup> in forestland. After forestland conversion to cropland, the annual cumulative N2O flux significantly increased by 76-491%. In the short-term and long-term croplands, tillage with fertilization induced cumulative soil N2O emissions that were 94% and 235% higher than those from tillage without fertilization. Fertilization contributed 63% and 84% to increased N2O emissions in the short-term and long-term croplands, respectively. A stepwise regression analysis showed that soil N2O emissions from croplands were mainly influenced by soil NO<sup>3</sup> <sup>−</sup> and NH<sup>4</sup> + availability and WFPS (water-filled pore space). Fertilization led to higher soil NH<sup>4</sup> <sup>+</sup> and NO<sup>3</sup> − concentrations, which thus resulted in larger N2O fluxes. Thus, to reduce soil N2O emissions and promote the sustainable development of the eco-environment, we recommend limiting the conversion of forestland to cropland, and meanwhile intensifying the shift from grain to green or applying advanced agricultural management practices as much as possible.

**Keywords:** land use change; tillage; fertilization; N2O fluxes; subtropical region

#### **1. Introduction**

Nitrous oxide (N2O) has been recognized as an important non-CO<sup>2</sup> greenhouse gas, with 298 times greater global warming potential than that of CO<sup>2</sup> based on a 100-year time horizon [1]. In the past 150 years, atmospheric N2O concentrations have greatly increased from 270 to 324 ppb [1], and the terrestrial biosphere is still a net source of atmospheric N2O [2]. Agricultural soils are the largest N2O source, contributing about 60% of global anthropogenic N2O emissions [1,3]. To guarantee a secure food supply for global human population growth, agriculture intensification might cause more N2O emission increases in the future [4,5]. Soil N2O is mainly produced by microbial nitrification and denitrification [6,7], and strongly affected by substrate availability (e.g., NH<sup>4</sup> <sup>+</sup> and NO<sup>3</sup> − concentrations) and environmental conditions (e.g., temperature and moisture) [6,8–11].

Many previous studies have proven that N2O emissions are greatly impacted by different land uses and management strategies [12,13]. In particular, the adjustment of land use is viewed as a crucial anthropogenic N2O emission source via its significant influences on N substrates and environmental conditions [1,9,12,13]. The existing studies focusing on the effects of land use change on N2O emissions mainly address rice paddy conversion to vegetable fields or citrus orchards [3,9,14], forest conversion to tea plantations [11,15], forest conversion to pasture or cropland [12], and cropland conversion to forestland [16,17]. Forestland converting to agricultural land has generally led to significant N2O emissions owing to deforestation and management practices (e.g., fertilizer application and tillage) [11–13,18]. In response to increasing socioeconomic development and demands for food, the specific land use change of forestland to cropland still occurs often around the world [11,12]. Although there have been a few N2O flux measurements since this type of conversion was published [12], the influence of forest conversion to cropland on N2O fluxes and their driving mechanisms are still not fully understood in many regions.

Tillage and fertilization are the most important management practices after the conversion of forestland to cropland; they significantly affect soil properties as well as soil carbon (C) and nitrogen (N) availability and ultimately regulate the production and emission of N2O [9–11,19,20]. However, tillage and fertilization have different driving mechanisms that affect N2O emissions. Tillage can influence the soil's physical structure (e.g., aggregation, bulk density and aeration), moisture, temperature, and C and N availability [19,21], which likely influence the microbial community and activity and thus N2O emissions [9,22]. Grandy and Robertson reported that in initial cultivation, tillage practices significantly disturbed the soil structure and released large amounts of organic C and N from macroaggregates, thus accelerating the soil N2O fluxes [18]. Fertilization has been widely understood to significantly increase N2O emissions, mainly through supplying more inorganic N concentrations and sufficient substrates (NH<sup>4</sup> <sup>+</sup> and NO<sup>3</sup> −) for nitrification and denitrification [8,11,23–25]. Bouwman et al. analyzed 139 N2O studies from agricultural regions and found that N2O emissions generally increase with N application rates, especially above the application rates of 100 kg N ha−<sup>1</sup> [26]. Recently, Chen et al. reported that forest conversion to tea fields significantly triggered substantial N2O emissions in the first year due to a high basal fertilizer input and intense tillage [11]. Previous studies have made great progress in explaining how tillage and fertilization practices drive N2O emissions in agricultural lands [10,19,24,26]. More studies are still needed to quantify the relative contributions of tillage and fertilization to increased N2O emissions when specifically converting forestland to cropland, which would be helpful for developing suggestions on how to reduce N2O emissions in the transformed croplands.

The Sichuan Basin is the largest agricultural region in Southwest China, accounting for 7% of the total cropland and supplying 10% of the total agricultural products of China [27]. In recent years, many studies have paid attention to the N2O fluxes from agricultural soils and obtained some significant observations in the Sichuan Basin [24,25,28,29]. However, previous studies mainly focused on the influence of fertilizer application (e.g., type and rate), which has resulted in agricultural lands having more N2O emissions compared to other land uses. Driven by afforestation policies and increasing food demand, land use conversion between forestland and cropland often occurs in this region. A recent study investigated the effects of afforestation on soil N2O emissions and found that afforestation significantly decreased N2O fluxes compared to those in cropland [17]. However, the effect of forestland conversion to cropland on soil N2O emissions and its underlying mechanisms are still uncertain in this region. Understanding the effects of tillage and fertilization practices on soil N2O emissions after forestland conversion to cropland will be beneficial for evaluating the environmental impacts of land use change on soil N2O emissions in the Sichuan Basin, and thus for suggesting appropriate technological approaches to mitigate these effects.

In this study, we measured N2O emissions and environmental variables from forest soils (as control), new croplands converted from forest (tillage with and without fertilization), and long-term croplands (tillage with and without fertilization) in the Sichuan Basin of Southwest China. The specific aims were (1) to quantify the influences of tillage and fertilization on soil N2O emissions after land use conversion of forestland to cropland, (2) to evaluate the short-term and long-term land use change impacts on soil N2O emissions, and (3) to identify the potential mechanisms driving the increased soil N2O emissions induced by tillage and fertilization after land use conversion.

## **2. Materials and Methods**

## *2.1. Study Area*

This study was carried out at the Yanting Agro-Ecological Station of Purple Soil of the Chinese Academy of Sciences (31◦160 N, 105◦270 E), located in the central Sichuan Basin of Southwest China with an altitude of 400 to 600 m [27]. It exhibits a moderate subtropical monsoon climate, with a mean annual precipitation and temperature of 836 mm and 17.3 ◦C (30-year mean). The widely distributed soil in the study region is called purple soil locally and is classified by the FAO soil classification as a Eutric Regosol and as a Pup-Orthic Entisol by the Chinese soil taxonomy [27]. The study area is an intensive agricultural production region in China. The sloping croplands have relatively thin soil thicknesses of 30–80 cm and slopes of 3–15%, and wheat (*Triticum aestivum* L.)-maize (*Zea mays* L.) rotation is the main cropping pattern in this region [17].

### *2.2. Experimental Design*

The selected forestland is dominated by *Cupressus funebris* with a mean diameter of 13.2 cm at breast height, a mean height of 16 m, and a density of 1595 stems ha−<sup>1</sup> [30]. In late July 2016, a portion of the forestland was cleared of trees and roots and then converted to cropland. To assess the short-term effects of management practices on soil N2O emissions, the newly converted croplands were cultivated from November 2016. In new croplands, two treatments of tillage—without fertilization (NC-T) and with fertilization (NC-TF)—were established with three replicates in a randomized block design (size 3 m × 3 m). The long-term croplands (tillage without fertilization, LC-T, and tillage with fertilization, LC-TF), which were adjacent to the newly converted croplands, were also established with three replicates in a randomized block design (size 4 m × 6 m). The long-term croplands converted from forestland have been cultivated since 2003. Moreover, the selected forestland was used as the control (CK) and had three replicate plots (size 3 m × 3 m). All the treatments had the same soil type (Regosols) and slope (5%).

Following the local cropping regimes, croplands were conventionally cultivated under a wheat-maize rotation system (winter wheat from November to May rotating with summer maize from June to October). Tillage practice involved conventional tillage with harrowing (approximately 20 cm deep) twice a year before sowing. The mineral N fertilization (urea) rates were 130 kg N ha−<sup>1</sup> and 150 kg N ha−<sup>1</sup> for wheat and maize with fertilization treatments, respectively [23]. All the fertilizers were manually applied and incorporated into the topsoil (0–20 cm) together with harrowing. Then, wheat and maize seeds were directly drilled into the soil. No irrigation was applied during the growth of either wheat or maize.
