*4.1. E*ff*ects of Land Use Conversion on Soil N2O Emissions and Yield-Scaled N2O Emissions*

In this study, the annual cumulative N2O flux from the forest was 0.21 kg N ha−<sup>1</sup> yr−<sup>1</sup> , which was significantly lower than the average annual N2O flux of the global forest (1.429 kg N ha−<sup>1</sup> yr−<sup>1</sup> ) [36]. Land use conversion from forestland to cropland significantly increased the annual cumulative N2O emissions by 76–491% in this study (Table 3). van Lent et al. reviewed the literature and reported that land use conversion from forest to cropland greatly increased the annual cumulative N2O emissions by 330% on average in the tropics and subtropics [13]. These results confirmed that land use change has a profound impact on soil N2O fluxes [9,11–14], specifically conversion of forestland to cropland. Some studies have proved that tillage and fertilization are the most important management practices increasing soil N2O emissions after land use conversion from forestland to cropland [11,19]. Tillage practices could significantly increase soil aeration conditions. On the one hand, good soil aeration condition enhanced soil organic matter mineralization [34], and subsequent nitrification [3], inducing higher N2O production. On the other hand, good soil aeration condition accelerated the gas exchange between soil and atmosphere, which could promote the diffusivity of N2O from soil into atmosphere [3,14]. Application of mineral N fertilizer could directly increase soil inorganic N concentrations (Figure 3), providing substrates for the two main N2O production processes of nitrification and denitrification [8,9,28]. In this study, the N2O flux from the croplands all showed pulse emissions following tillage and N fertilization events (Figure 4a), thus resulted in much greater cumulative N2O emissions compared to the forestland (Table 3).

Previous studies regarding the effects of forestland conversion to cropland on soil N2O emissions mainly focused on long-term cultivation croplands [12], but few studies have measured N2O emissions from recently transitioned croplands. Our study indicated that forestland conversion to cropland in the first year significantly increased annual cumulative N2O emissions by 124% and 334% (Table 3). Comparably, after two years of conversion of tropical forest to agriculture in French Guiana, the annual cumulative N2O emissions from fertilized croplands increased by 90% [12]. Obviously, the increased extent of N2O emissions in our study was higher than that in the study by Petitjean et al. [12]. This difference might be attributed to the differences in climate, soil properties and amount of N fertilizer [26]. It is remarkable that the high N2O emissions in the initial stage after land use conversion should not be ignored.

In agricultural systems, yield-scaled N2O emissions were widely used as a metric of the important global challenge for guaranteeing food security and reducing N2O emissions [37,38]. In our study, the yield-scaled N2O emissions were 0.16 and 0.19 kg N Mg−<sup>1</sup> grain in croplands after long-term cultivation (Table 3), which were similar to the results reported by Bayer et al. [20] and Tang et al. [39] in other subtropical regions. However, the yield-scaled N2O emissions in the new croplands were 21% and 106% higher than that in the long-term converted croplands (Table 3). These results further highlighted the importance of monitoring N2O emissions at the initial stage after land use conversion [3,11]. In the newly converted croplands, the lower crop yields mean a lower uptake of available N by plant, leaving more available N for nitrification and denitrification processes, which promoted the production of N2O. After long-term cultivation, the capacity of plant N uptake was enhanced, inducing a relative lower ratio of available N for N2O production [10,40]. Furthermore, after long-term cultivation in the present study, the increase extent of crop yield was 64% and 66% compared to the new croplands, which were significantly higher than the increase extent of N2O emissions (−21% and 37%), inducing significant decreases of the yield-scaled N2O emissions. Tillage with fertilization practices resulted in yield-scaled N2O emissions 19% higher than only tillage practice in the long-term converted croplands (Table 3). It is obvious that fertilization has induced a much higher increase extent of N2O emissions than that of crop yield. Previous studies showed that the N fertilizer application rate was one of the key factors influencing soil N2O production [26]. Meanwhile, N fertilizer application rate also had considerable impacts on crop yield [41]. Therefore, it is very important to determine a reasonable

rate of N fertilizer application in the Sichuan Basin after forestland conversion to cropland, which can simultaneously achieve high yield and mitigate soil N2O emission. In addition, the N fertilizer type (e.g., mineral, organic, or mix of mineral and organic), application time and proportion (e.g., all as base fertilizer, or part as base and other as topdressing), as well as application depth (surface or deep) would also influence N fertilizer efficiency and N2O emission [41].

Overall, land use conversion from forestland to cropland and subsequent tillage and fertilization practices significantly increased the cumulative soil N2O emissions. In particular, the yield-scaled N2O emissions were significantly higher in the newly converted croplands than in the long-term converted croplands. These results further indicate that the effect of land use conversion on soil N2O emissions should not be ignored in the initial years after conversion. Therefore, we strongly recommend limiting the conversion of forestland to cropland as much as possible.

## *4.2. Factors Regulating the Increased Soil N2O Emissions Induced by Tillage and Fertilization*

In this study, tillage in the croplands increased the cumulative N2O emissions by 124% and 76% compared to those from forestland in the short-term and long-term after conversion, respectively (Table 3). This result indicates that tillage could induce significant increases in soil N2O emissions after land use conversion. The statistical analyses showed that tillage significantly decreased the soil bulk density (Table 1), and thus increased soil porosity, while also decreasing the proportion of soil water-stable macroaggregates and MWD by 49% and 61% on average compared to those in forestland (Table 2). The breakage of soil macroaggregates releases more soil organic matter and increases N availability [18,21,34]. This phenomenon was observed through the decrease in the average SOC and TN contents by 49% and 34%, respectively, in croplands under only tillage compared to the contents in forestland (Table 1). Tillage physically disturbs the soil structure and increases soil aeration [34,42], and therefore enhances soil organic N mineralization and microbial activities [3,11,19]. In this study, cropland tillage induced small NH<sup>4</sup> <sup>+</sup> peaks in the initial period of each cropping season that lasted for approximately three weeks (Figure 3a), thus likely increasing the NH<sup>4</sup> <sup>+</sup> to NO<sup>3</sup> − transformation rate. This possibility was further supported by the higher NO<sup>3</sup> − concentrations in the tillage treatments compared to those in the forestland (Figure 3b). The increase in soil NO<sup>3</sup> − would further enhance the denitrification rate [17,23]. Consequently, tillage in the croplands increased soil aeration and promoted soil organic N mineralization, thus inducing N2O pulse emissions that did not occur in the forestland and accounted for approximately 52% of the annual cumulative N2O flux during the experimental period (Figure 4).

Tillage with fertilization significantly increased the cumulative soil N2O emissions by 94% and 235% compared to tillage without fertilization in this study (Table 3). In another subtropical region of China, a previous study found that soil N2O emissions were 405% higher on average in a conventional fertilizer treatment than in a tillage alone treatment after land use conversion [11]. In the current study, tillage with fertilization induced much higher N2O pulse emissions, mainly due to the rapid increases in soil NH<sup>4</sup> <sup>+</sup> and NO<sup>3</sup> − concentrations after mineral N fertilizer application to the croplands (Figure 3a,b). The high NH<sup>4</sup> <sup>+</sup> and NO<sup>3</sup> − concentrations following fertilization directly provide sufficient substrates for nitrification and denitrification and thus significantly stimulate soil N2O emissions [8,11,20,23]. This phenomenon has been reported in many previous studies on different croplands [5,10,43]. On average, pulse N2O emissions following fertilization accounted for 67% of the annual cumulative N2O fluxes (Figure 4). Previous study reported that N2O emission pulses following N fertilization contributed to approximately 70% of the annual cumulative N2O fluxes from croplands in the same study area [17].

In the newly converted cropland, 37% and 63% of the increased soil N2O emissions could be attributed to tillage and fertilization, respectively, while in the long-term cropland, the corresponding rates were 16% and 84%. This result indicates that fertilization had a much greater effect on increasing soil N2O emissions than tillage after forestland conversion to cropland in the Sichuan Basin, and the effect tended to be larger several years after conversion. Long-term repeated tillage significantly reduced soil structural stability and decreased organic matter (Tables 1 and 2), resulting in a lower rate of soil organic N mineralization [18–20,22]. Long-term mineral N fertilization stimulated the activity and abundance of soil ammonia-oxidizing bacteria in the study area [8], which could promote nitrification and subsequent denitrification [23].

The stepwise regression analysis results indicated that soil WFPS and temperature could explain 82% of the variations in N2O fluxes from forestland (Table 4). Soil temperature and moisture are important factors that influence soil N2O emissions by affecting microbial activities [8,23,44]. The low soil moisture and temperature during the winter wheat season (Figure 2) generally inhibited microbial activities, resulting in low N2O emissions [3,11]. However, the stepwise regression analysis showed that soil N2O emissions from the croplands were mainly influenced by soil NO<sup>3</sup> <sup>−</sup> and NH<sup>4</sup> <sup>+</sup> availability and WFPS, which explained 78-90% of the variations in N2O fluxes (Table 4). This result implies that after the conversion of forestland to cropland, the primary factors regulating soil N2O emissions were soil NO<sup>3</sup> <sup>−</sup> and NH<sup>4</sup> <sup>+</sup> availability. In the present study, we found the N2O emissions significantly correlated to soil NH<sup>4</sup> <sup>+</sup> and NO<sup>3</sup> <sup>−</sup> concentrations in the croplands (Figure 5f,h). The mean soil NH<sup>4</sup> + and NO<sup>3</sup> − concentrations in the tillage with fertilization treatments were 7.7 and 3.7 times greater, on average, than those in the tillage without fertilization treatments. Therefore, after the conversion of forestland to cropland, fertilization only increased soil N2O emissions to a greater degree than tillage, which was mainly attributed to the higher soil inorganic N levels caused by fertilization.

Additionally, the soil DOC concentration significantly decreased after the conversion of forestland to cropland (Figure 3c). The lower level of soil DOC with higher N2O fluxes in the croplands compared to the forestland indicated the negative relationship between soil DOC concentration and N2O fluxes induced by land use conversion (Figure 5j). Several studies have shown that soil DOC is an important factor influencing N2O emissions [9,11,17,44]. The substantial soil DOC availability in the forestland may favor complete denitrification and thus decrease N2O fluxes [17]. Following land use conversion, the decrease in soil DOC concentration could enhance incomplete denitrification, thereby increasing N2O fluxes [44]. In the short-term after conversion, the combination of soil DOC and NH<sup>4</sup> <sup>+</sup> availability and soil WFPS explained 74% of the variation in N2O fluxes from the tillage and fertilization treatment (Table 4).

Applying crop residues with a high C/N ratio (such as maize) combined with synthetic N fertilizer could be an optimal strategy for mitigating N2O emissions in the study area [28,29]. Moreover, in recent years, researchers have developed many new technological approaches to maintain the sustainable development of the agricultural ecosystem. For example, a new Biogeosystem Technique methodology reported by Kalinitchenko et al. [45,46] could improve soil aggregate structure, promote soil organic matter synthesis and reservation, and thus likely reduce greenhouse gas production. Therefore, after the conversion of forestland to cropland in the Sichuan Basin, to guarantee the sustainable food supply and ecosystem development, we also recommend application of mineral fertilizer and crop residues (such as maize), or the adoption of advanced management technologies as potential measures for conserving N in destroyed forest and mitigating N2O emissions in the transformed croplands.

#### **5. Conclusions**

This study found that the annual cumulative N2O flux was 0.21 kg N ha−<sup>1</sup> yr−<sup>1</sup> in the forestland, which significantly increased by 76–491% after land use conversion of forestland to cropland in the Sichuan Basin. In the short-term and long-term croplands, fertilization contributed 63% and 84%, while tillage contributed 37% and 16% to the increased soil N2O emissions, respectively. Fertilization exhibited a much greater effect on increasing soil N2O emissions than tillage after the conversion of forestland to cropland, and the effect tended to be stronger several years after conversion. After forestland conversion to cropland, the soil N2O emissions were mainly regulated by soil NO<sup>3</sup> − and NH<sup>4</sup> <sup>+</sup> availability. The direct land use conversion without any scientific management practices significantly influenced soil properties and thus stimulated N2O emissions. Tillage disturbed soil structure, decreased bulk density and increased soil aeration in the croplands, thus enhancing soil

organic N mineralization, while the application of mineral N fertilizer directly led to rapid increases in soil NH<sup>4</sup> <sup>+</sup> and NO<sup>3</sup> −concentrations. Tillage and fertilization induced increases in the inorganic N concentration to different extents, which thus resulted in different magnitudes of N2O fluxes. This study primarily suggests limiting the conversion of forestland to cropland in the Sichuan Basin as much as possible. Moreover, we recommend intensifying the shift from grain to green on the premise of ensuring the supply of grain production, and also recommend the adoption of technological approaches to mitigate N2O emissions in both the destroyed forest and transformed croplands for the sustainable development of the ecosystem.

**Author Contributions:** Conceptualization, X.R. and B.Z.; data curation, X.R.; methodology, X.R., H.B., S.T.R., and B.Z.; writing—original draft, X.R.; writing—review and editing, X.R. and B.Z. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Key Research and Development Program of China (Grant No. 2017YFD0200105).

**Acknowledgments:** The members of the Yanting Agro-Ecological Station of Purple Soil of the Chinese Academy of Sciences are kindly thanked for providing weather data and technical help. The authors greatly appreciated the constructive comments from the reviewers, which greatly helped improve this manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
