This is the accepted version of the article: Wang, Chun; Li, Xiang; Min, Qingwen; [et al.]. Responses of greenhouse-gas emissions to land-use change from rice to jasmine production in subtropical China. DOI 10.1016/j.atmosenv.2018.12.032 This version is avaible at https://ddd.uab.cat/record/203308 under the terms of the license 1 Responses of greenhouse-gas emissions to land-use change from 2 rice to jasmine production in subtropical China 3 4 Chun Wanga,b, Qingwen Minc, Weiqi Wanga,d,e,f, *, Jordi Sardanse,f,*, Congsheng 5 Zenga,d, Chuan Tonga,d, Josep Peñuelase,f a 6 Institute of Geography, Fujian Normal University, Fuzhou 350007, China b 7 Key Laboratory of Wetland Ecology and Environment, Northeast Institute of Geography and 8 Agroecology, Chinese Academy of Sciences, Changchun 130102, China c 9 Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of 10 Sciences, Beijing 100101, China d 11 Key Laboratory of Humid Subtropical Eco-geographical Process, Ministry of Education, 12 Fujian Normal University, Fuzhou 350007, China e 13 CSIC, Global Ecology Unit CREAF-CSIC-UAB. 08913 Bellaterra, Catalonia, Spain f 14 CREAF. 08913 Cerdanyola del Vallès, Catalonia, Spain 15 16 17 *Corresponding author at: Institute of Geography, Fujian Normal University, Fuzhou 350007, 18 China. Tel.: +86 591 83465214. 19 E-mail addresses: wangweiqi15@163.com 20 E-mail addresses: j.sardans@creaf.uab.cat 21 22 23 24 25 26 27 28 1 29 ABSTRACT 30 We studied the impacts of an increasingly common anthropogenic change in land use from 31 paddy field to jasmine fields on the emission of greenhouse gases (GHGs), which have 32 supposed the transformation of more than 1200 ha only in the last decade in the surroundings 33 of Fuzhou city in response to economic changes. The possible increases that this can suppose 34 constitutes and environmental concern in China. We studied areas dedicated to rice crop that 35 have been partially converted to jasmine cultivation with some parts still kept as rice fields. 36 Emissions of CO2, CH4 and N2O varied significantly among the seasons. CO2 and CH4 37 cumulative emissions and the global-warming potential (GWP) of these emissions were 38 significantly lower in the jasmine than the paddy field. N2O emission, N2O cumulative 39 emission, however, were higher in the jasmine than the paddy field, despite in some concrete 40 studied periods the differences were not statistically significant. The total decrease in GHG 41 emissions from the conversion from rice to jasmine production was strongly influenced by the 42 indirect effects of various changes in soil conditions. The expected changes due to the great 43 differences in water and fertilization use and management and organic matter input to soil 44 between these two crops were in great part due to modified soil traits. According to structural 45 equation models, the strong direct effects of the change from rice to Jasmine crop reducing 46 the emissions of CO2 and N2O were partially decreased by the indirect effects of crop type 47 change decreasing soil pH and soil [Fe2+] for CO2 emissions and by decreasing soil salinity 48 and soil [Fe3+] for N2O emissions. The negative effects of the crop conversion on CH4 49 emissions were mostly due to the globally negative indirect effects on soil conditions, by 50 decreases in soil salinity, water content and [Fe2+]. Soil salinity, water content, pH, [Fe2+], 51 [Fe3+] and [total Fe] were significantly lower in the jasmine than the paddy field, but 52 temperature had the opposite pattern. CO2 emissions were generally correlated positively with 53 salinity, temperature, and water content and negatively with [Fe3+] and [total Fe] in both fields. 54 CH4 emissions were positively correlated with salinity, temperature, water content and pH in 55 both fields. N2O emissions were positively correlated with temperature and were negatively 56 correlated with water content, pH, [Fe2+], [Fe3+] and [total Fe] in both fields. CO2 was the 57 most important GHG for the GWPs, and the total GWP was significantly lower for the 58 jasmine than for the rice cropland field. The change in the land use in this area of paddy fields 2 59 will decreased the global GHG emission, and the effect on the GWPs was mostly due to 60 changes in soil properties. 61 62 Keywords: CO2, CH4, N2O, emission, GWP, paddy field, jasmine cultivation 63 64 1. Introduction 65 Rice is the most important cereal crop for more than half of the world’s population. The Food 66 and Agricultural Organization of the United Nations (2009) estimated that rice production 67 must increase by 40% by the end of 2030 to meet the rising demand from the increasing 68 population. Rice cultivation, however, needs more water and labor and provides a lower 69 income than other crops, such as groundnuts, sweet potatoes, and coffee (Bua and Ojirot, 70 2014). Land-use conversion, particularly within cropland, is common and driven by the 71 market’s economy (Houghton et al., 1999). Some paddy fields have been converted to 72 vegetables in response to economic changes and the need for environmental conservation. The 73 conversion of rice paddy fields to the production of other crop species frequently changes soil 74 texture and microbial communities (Dai et al., 2016), soil aggregates and organic-carbon 75 concentrations (Wang et al., 2014), methane emissions (Hu et al., 2016), plant diversity of 76 rural herbs (Wu et al., 2016) soil bulk density and porosity (Li et al., 2017) and the soil carbon 77 and nitrogen concentrations (Huang et al., 2009).. Paddy fields, though, are very important 78 sources of greenhouse gases (GHGs), especially methane (CH4) and nitrous oxide (N2O) 79 (Myhre et al., 2013), so minimizing the release of these very potent GHGs from paddies is a 80 key aim for better management strategies to mitigate the adverse impacts on climate change 81 by Chinese cropland activity as a whole. China has the second largest area of rice cultivation 82 in the world, and the associated GHG emissions account for about 40% of the total 83 agricultural sources of GHGs. Moreover, 90% of the paddies in China are in the subtropics, 84 such as in Fujian, Jiangxi and Hunan Provinces. 85 Jasmine tea is unique, and China is the only country that has mastered the critical 86 scenting technologies. Protecting this production system is thus important for the protection 87 and inheritance of Chinese culture and traditional technologies. More than half of the jasmine 3 88 tea in China is produced in Fuzhou Province (Xu et al., 2001; Yang et al., 2008; Xu, 2012). 89 The system for culturing jasmine and other tea plants near the city of Fuzhou was added in 90 2014 to the Globally Important Agricultural Heritage Systems due to its long historic, 91 ecological and cultural function in this region (Lin et al., 2014; Ren et al., 2015). The land 92 area devoted to jasmine production is currently increasing due to the great economic benefits, 93 mainly by conversion from rice cropland, but little is known about the quantitative changes in 94 emissions and the possible changes in soil properties underlying this change in cropland 95 activity. We studied the impacts of an increasingly common change in land use from paddy 96 field to jasmine fields on the emission of greenhouse gases (GHGs), which had supposed the 97 transformation of more than 1200 ha only in the last decade in the surroundings of Fuzhou 98 city in response to economic changes. The possible impacts constitute and environmental 99 concern in China. 100 To increase our understanding of the effects of land-use changes from rice to jasmine 101 production on GHGs emissions, we: (1) identified the changes in soil properties associated 102 with the conversion of rice to jasmine production, (2) measured GHG emissions and 103 estimated global-warming potentials (GWPs) and (3) identified the mechanistic relationships 104 among the shifts in GHG emissions and soil properties. 105 106 2. Material and methods 107 2.1. Study area and experimental fields 108 This study was conducted in the Changshan district of Fuzhou, Fujian province (China, Fig. 1). 109 The climate is subtropical, with mean annual temperatures and precipitation of 19.7 ºC and 110 1348.8 mm, respectively (Wang et al., 2015a). The experimental period was from April 2015 111 to March 2016. During this period there were two rice seasons, early paddy season (from 16 112 April to 16 July), with transplantation at 16 April, and the late paddy season (from 25 July to 113 6 November), with transplantation on 25 July. The experiment was conducted at two sites, one 114 a paddy field, and the other a jasmine field, a field before and after conversion from rice to 115 jasmine production. 116 The soil of the rice paddy field was poorly drained, and the proportions of sand, silt and 117 clay particles in the top soil (15 cm surface soil) were 28, 60 and 12%, respectively (Wang et 4 118 al., 2016a). Other surface-soil properties were: bulk density 1.1 g cm-3, pH (1:5 soil:H2O) 6.5, 119 organic carbon 18.1 g kg−1, total nitrogen (N) 1.2 g kg-1 and total phosphorus (P) 1.1 g kg-1 120 (Wang et al., 2015b, c). The water level was maintained at 5-7 cm above the soil surface from 121 0 to 37 days after transplantation (DAT) by an automatic water-level controller, and the water 122 was then drained between 37 and 44 DAT. The soils in paddy field were kept moist between 123 44 and 77 DAT, and the paddy field was subsequently drained two weeks before harvest. The 124 paddy field was plowed to a depth of 15 cm with a moldboard plow and leveled two days 125 before rice transplantation. Mineral fertilizers were applied in three batches as complete 126 (N-P2O5-K2O at 16-16-16%; Keda Fertilizer Co., Ltd. Jingzhou, China) and urea (46% N) 127 fertilizers. Fertilizers were first applied one day before transplantation at rates of 42 kg N ha-1, 128 40 kg P2O5 ha-1 and 40 kg K2O ha-1. A second application was done during the tiller-initiation 129 stage (7 DAT) at rates of 35 kg N ha-1, 20 kg P2O5 ha-1 and 20 kg K2O ha-1 and a third 130 application was done during the panicle-initiation stage (56 DAT) at rates of 18 kg N ha-1, 10 131 kg P2O5 ha-1 and 10 kg K2O ha-1. 132 A nearby paddy field, with the same rice crop history and soil traits was converted to 133 jasmine production about seven years ago. The jasmine growing season is from April to 134 October. Currently the soil in the jasmine field contained 25, 59 and 16% sand, silt and clay, 135 respectively. The soil had a bulk density of 1.2 g cm-3, pH of 4.4, salinity of 0.15 mS cm-1 and 136 concentrations of total carbon, total N, total P and total potassium of 11.7, 1.1, 0.5 and 13.3 g 137 kg-1, respectively. The jasmine was cultivated using a ridge and ditch system, with a ridge 138 height of 20 cm, ridge width of 100 cm and ditch width of 30 cm. Jasmine branches 10 cm 139 long were transplanted by hand in the ridges at a spacing of 3(width)×20 (longth) cm in April 140 2008 and have grown for seven years. The jasmine field was not plowed, but the soil was 141 ridged each year after the jasmine was cut. Jasmine branches and leaves began to grow from 142 early April to early May. Budding and infancy were from early May to the end of May. 143 Flowering was from early June to the end of September, when the final growth period began. 144 A complete fertilizer (N:P2O5:K2O=16:16:16%) was applied in two unequal splits. Fertilizers 145 were first applied after the jasmine was cut at rates of 20 kg N ha-1, 20 kg P2O5 ha-1 and 20 kg 146 K2O ha-1. The second application was done one day after the first jasmine flowers were 147 collected, at rates of 16 kg N ha-1, 16 kg P2O5 ha-1 and 16 kg K2O ha-1. 5 148 149 2.2. Measurement of CO2, CH4 and N2O emissions 150 Static closed chambers were used to measure soil CO2, CH4 and N2O emissions, as described 151 by Wang et al. (2015b). The chambers were made of rigid PVC and consisted of two parts, an 152 upper opaque compartment (100 cm height, 30 cm width, 30 cm length) placed on a 153 permanently installed bottom collar (10 cm height, 30 cm width, 30 cm length). Each 154 chamber had two battery-operated fans to mix the air inside the chamber headspace, an 155 internal thermometer to monitor temperature changes during gas sampling, a gas-sampling 156 port with a neoprene rubber septum at the top of the chamber for collecting gas samples from 157 the headspace and a vent to prevent pressure buildup. Three replicate chambers were used in 158 each crop type. In each sampling time we collected 6 samples for each variable (2 crop types 159 x 3 plots = 6) 160 Gas flux was measured for 30 min for all chambers once a week during the rice growing 161 season, twice a week during the jasmine growing season and once a month during the other 162 seasons. The temperature in the chamber did not vary significantly during the sampling. Gas 163 samples were collected from the chamber headspace using a 100-ml plastic syringe with a 164 three-way stopcock 0, 15 and 30 min after chamber deployment. The samples were 165 immediately transferred to 100-ml air-evacuated aluminum-foil bags (Delin Gas Packaging 166 Co., Ltd., Dalian, China), sealed with butyl rubber septa and transported immediately to the 167 laboratory for the analysis of CO2, CH4 and N2O. 168 CO2, CH4 and N2O concentrations in the headspace air samples were determined by gas 169 chromatography (Shimadzu GC-2010 and Shimadzu GC-2014, Kyoto, Japan) using a 170 stainless steel Porapak Q column (2 m length, 4 mm OD, 80/100 mesh). A 171 methane-conversion furnace, flame ionization detector (FID) and electron-capture detector 172 (ECD) were used for the determination of the CO2, CH4 and N2O concentrations, respectively. 173 The operating temperatures of the column, injector and detector for the determination of the 174 CO2, CH4 and N2O concentrations were adjusted to 45, 100 and 280 °C; to 70, 200 and 175 200 °C and to 70, 200 and 320 °C, respectively. Helium (99.999% purity) was used as a 176 carrier gas (30 ml min-1), and a make-up gas (95% argon and 5% CH4) was used for the ECD. 177 The gas chromatograph was calibrated before and after each set of measurements using 503, 6 178 1030 and 2980 μl CO2 l-1 in He; 1.01, 7.99 and 50.5 μl CH4 l-1 in He and 0.2, 0.6 and 1.0 μl 179 N2O l-1 in He (CRM/RM Information Center of China) as standards. Three injections were 180 used for each analysis. One sample was injected into the gas chromatograph for each analysis. 181 The detection limits of the instrument for CO2, CH4 and N2O were 1, 0.1 and 0.05 ppm, 182 respectively. CO2, CH4 and N2O fluxes were then calculated as the rate of change in the mass 183 of CO2, CH4 and N2O per unit of surface area and per unit of time. The cumulative emissions 184 were calculated by multiplying the measured emissions in one hour in each sampling day by 185 24 hours and then the obtained value multiplied by the half of the days from the previous 186 measurement day plus the half of the days until the next measurement day. 187 188 2.3. GWP 189 GWP is typically estimated using CO2 as the reference gas, and a change in the emission of 190 CH4 or N2O is converted to “CO2-equivalents”. The GWP for CH4 is 34 (based on a 100-year 191 time horizon and a GWP for CO2 of 1), and the GWP for N2O is 298 (Myhre et al., 2013). The 192 GWP of the combined emission of CH4 and N2O was calculated as: 193 GWP = (cumulative CO2 emission × 1) + (cumulative CH4 emission × 34) + (cumulative 194 N2O emission × 298) 195 2.4. Measurement of soil properties 196 Three replicate soil samples were collected from the paddy and jasmine fields in each 197 sampling moment. The samples were transported to the laboratory and stored at 4 °C until 198 analysis. The temperature, pH, salinity and water content of the top 15 cm of soil were 199 measured in situ at each plot on each sampling day. Temperature and pH were measured with a 200 pH/temperature meter (IQ Scientific Instruments, Carlsbad, USA), salinity was measured using 201 a 2265FS EC meter (Spectrum Technologies Inc., Paxinos, USA) and water content was 202 measured using a TDR 300 meter (Spectrum Field Scout Inc., Aurora, USA). Soil samples were 203 collected from the 0-15 cm layer from each plot for the determination of ferric, ferrous and total 204 Fe concentrations. Total Fe concentration was determined by digesting fresh soil samples with 205 1M HCl. Ferrous ions were extracted using 1,10-phenanthroline and measured 206 spectrometrically (Lu, 1999; Wang et al., 2014b). Ferric concentration was calculated by 207 subtracting the ferrous concentration from the total Fe concentration. 7 208 209 2.5. Statistical analysis 210 Differences in soil properties and CO2, CH4 and N2O emissions between the land uses were 211 tested for statistical significance by general mixed models and by generalized mixed models if 212 a variable was non-normally distributed, using plot as a random factor and time as nested 213 factor within plot as random independent factors when time was included in the analysis. We 214 used the “nlme” (Pinheiro et al., 2016) R package with the “lme” function. We chose the best 215 model for each dependent variable using Akaike information criteria. We used the MuMIn 216 (Barton, 2012) R package in the mixed models to estimate the percentage of the variance 217 explained by the model. 218 The averaged soil properties and CO2, CH4 and N2O cumulative emissions, each GWP 219 and total GWP were tested for statistical significance as described above, with plot as a 220 random factor. Statistical significance of the differences among emissions and soil variables 221 throughout time were tested by repeated-measures analyses of variance. The relationships 222 between mean GHG emissions and soil properties were determined by Pearson correlation 223 analysis. These statistical analyses were performed using SPSS Statistics 18.0 (SPSS Inc., 224 Chicago, USA). 225 We used structural modelling to analyse the factors explaining the maximum variance of 226 the CO2, CH4 and N2O emissions throughout the study period as a function of cultivation type 227 and soil traits. This analysis provided information on how in this case the change in crop type 228 and the related changes in soil types were related among them in the form that explained most 229 variance of each studied gas emissions. In other words, we thus analysed how total effects of 230 crop type change on gas emissions are mediated throughout the indirect effects of crop type 231 on related soil traits shifts. This method elucidates the part of the relationships of the change 232 of crop with the shifts of soil gas emissions due to the corresponding shifts in the studied soil 233 variables. We fit the models using the sem R package (Fox et al., 2013) and acquired the most 234 parsimonious model using the Akaike information criterion. 235 236 3. Results 237 3.1. CO2 emissions from the paddy and jasmine fields 8 238 CO2 emission generally varied significantly among sampling dates and between fields 239 (P<0.01, Table 1, Fig. 2). CO2 flux in the paddy field was generally higher from April to 240 December (rice growth period, and the beginning of straw return for December, >264 mg m-2 241 h-1) and lower from January to March (fallow period, <100 mg m-2 h-1). CO2 flux in the 242 jasmine field was generally higher from April to August (jasmine rapid-growth period, >770 243 mg m-2 h-1) and lower from September to March of the next year (jasmine slow-growth period, 244 <300 mg m-2 h-1) (Fig. 2). Cumulative CO2 emission was lower in the jasmine than the paddy 245 field (P<0.05, Table 2). CO2 was the most important GHG for the GWPs, and total GWP was 246 significantly lower in the paddy than the jasmine field (P<0.05). 247 248 3.2. CH4 emissions from the paddy and jasmine fields 249 CH4 emission generally varied significantly among sampling dates and between fields 250 (P<0.01, Table 1, Fig. 2). CH4 flux in the paddy field was generally higher from April to 251 August (early growth period and the beginning of late growth period, >0.9 mg m-2 h-1) than 252 later (<0.7 mg m-2 h-1). CH4 flux was generally lower (<0.5 mg m-2 h-1) in the jasmine field 253 (Fig. 2). Cumulative CH4 emissions were lower in the jasmine than the paddy field (P<0.05). 254 255 3.3. N2O emissions from the paddy and jasmine fields 256 N2O emission generally varied significantly among sampling dates and between fields 257 (P<0.01, Table 1, Fig. 2). N2O flux was generally lower (<200 μg m-2 h-1) in the paddy field. 258 N2O flux in the jasmine field was generally higher from April to August (jasmine 259 rapid-growth period, >300 μg m-2 h-1) and lower from September to March of the next year 260 (jasmine slow-growth period, <105 μg m-2 h-1) (Fig. 2). Cumulative N2O emissions were 261 higher in the jasmine field (P<0.05, Table 2). 262 263 3.4. Differences in soil properties between the paddy and jasmine fields 264 Soil salinity, temperature, water content, pH and ferrous, ferric and total Fe concentrations 265 varied among the sampling dates, fields and interactions of sampling dates and fields (P<0.05; 266 Table 3, Fig. 3). Soil salinity, water content, pH, [Fe2+], [Fe3+] and [total Fe] were lower in the 9 267 jasmine than the paddy field (P<0.05), but temperature was significantly higher in the jasmine 268 than in the paddy field (P<0.05, Table 4). 269 270 3.5. Relationships between CO2, CH4 and N2O emissions and soil properties 271 The seasonal variation of CO2 emissions in the soils of the paddy fields was generally 272 correlated positively with salinity, water content (P<0.05; Table 5), and temperature (P<0.01), 273 and negatively correlated with soil pH (P<0.05; Table 5), and [Fe2+], [Fe3+], [total Fe] 274 (P<0.01). CH4 emissions were positively correlated with salinity, temperature, water content 275 (P<0.01), negatively correlated with [Fe2+] (P<0.05), and [Fe3+], [total Fe] (P<0.01). N2O 276 emissions were positively correlated with [Fe2+], [Fe3+], [total Fe] (P<0.05), and negatively 277 correlated with salinity, and water content (P<0.05). 278 In Jasmine cropland soils the seasonal variation of CO2 emissions in the soils of the 279 jasmine cropland was generally positively correlated with water content (P<0.05; Table 5), 280 and salinity, temperature (P<0.01), and negatively correlated with soil pH, [Fe 3+], [total Fe] 281 (P<0.01). CH4 emissions were not significantly correlated with the environmental factors. 282 N2O emissions were positively correlated with salinity, and temperature (P<0.01), and 283 negatively correlated with pH, [Fe3+], and [total Fe] (P<0.01). 284 285 3.6. SEM analyses 286 The total decrease in GHG emissions in the conversion from rice to jasmine production was 287 strongly influenced by the indirect effects of changing soil conditions. The strong indirect 288 effects of the change in crop type came from the decrease in soil pH for CO 2 and N2O, from 289 the decrease in soil [Fe2+] for CO2 and from the decrease in soil salinity and [Fe3+] for N2O 290 (Figures 4 and 5). These indirect effects were only partially counteracted by a positive direct 291 effect of the change of crop type. The increases in cumulative N2O emissions due to indirect 10 292 effects, that were higher in magnitude than their direct negative effects resulted in increased 293 emissions after the transition (Figures 4 and 5). The negative indirect effects of crop type on 294 CH4 emissions by lowering soil salinity, water content and [Fe2+] in the jasmine field were 295 only partially balanced by a positive direct effect, so the global effect was a decrease in CH4 296 emissions (Figures 4 and 5). 297 298 4. Discussion 299 4.1. Effects of changing from rice to jasmine production on soil properties 300 The soil properties of paddies were significantly different from those in jasmine crops. Rice 301 crops have a period of flooding that jasmine crops do not have. Soil salinity and pH were 302 lower in the jasmine than the paddy field, consistent with previous results (Yang and Zhang, 303 2014; Gaddanakeri et al., 2008). The higher salinity and pH in paddy resulted because about 304 half year was flooded. Thus, many ions, such as K+, Ca2+, Na+, Cl- are then dissolved in the 305 water, thereby incrementing the soil salinity and pH due to Fe3+ reduction consuming H+ 306 (Gribsholt et al., 2003). Moreover, the paddy receives higher fertilizer amounts, and the 307 chemical fertilizer rich in K+, which will also increase the soil salinity and pH. Furthermore, 308 irrigation water in paddy field is from rivers rich in ions such as K+, Ca2+, Na+, and Cl- 309 coming from seawater by tide, thereby increasing the soil salinity and pH (Sharma et al., 310 2012), whereas Jasmine crops in Fujian are only irrigated in very dry years, very infrequently. 311 Temperature, however, was significantly higher in the jasmine than in the paddy field, 312 specially for the paddy field flood period, likely due to the buffering effect of flooding on 313 temperature in the paddy field. Soil active [Fe2+], [Fe3+] and [total Fe] were lower in the 314 jasmine than in the paddy field, very probably as a result of the lower fertilizer application in 315 the jasmine field with respect to rice crop. There are two reasons: fertilizer increases the iron 316 reduction and oxidation by microbes (Gudzic et al., 2015) and alternation of wet and dry 317 periods in the paddy drives the non-active Fe change to active Fe, thereby increasing the 318 active Fe concentration (Fimmen et al., 2008). 319 . 11 320 321 4.2. Effects of changing from rice to jasmine production on CO2 emissions 322 CO2 emissions varied significantly among the seasons. CO2 emission varied seasonally, 323 increasing with crop growth and temperature. This pattern has already been reported by Liu 324 et al. (2011). Temperature controls CO 2 production and emission (Emmett et al., 2004; 325 Asensio et al., 2007; Inglett et al., 2012; Treat et al., 2014) by not only increasing soil 326 microbial activity (Vogel et al., 2014), but also by altering plant respiration (Atkin and 327 Tjoelker, 2003; Slot et al., 2013). Besides temperature and crop growth, CO2 emissions were 328 correlated positively with salinity and water content and negatively with [Fe3+] and [total Fe] 329 in both fields. In both croplands, the salinity is very low, but the salinity increases can provide 330 nutrients, thereby promoting the CO2 production and emissions. The jasmine field or the non 331 rice growth period of the paddy are dry, and when the water content increases there is an 332 increase in the CO2 production and emissions (Linn and Doran 1984). The CO2 emission was 333 negatively correlated with [Fe3+] and [total Fe] in both fields. Soil Fe3+ concentration 334 increment enhances the formation of iron plaque on the crop roots and thus limits the CO2 335 transport to the atmosphere (Butterbach-Bahl et al., 1997; Wassmann and Aulak, 2000). 336 Moreover, when soil Fe, Fe3+ concentrations increase, the rate of Fe3+ reduction also 337 increases, and reduced Fe 2+ accumulates in the soil (Wang et al., 2015c), which could 338 inhibit microbial activity (Wu et al., 2012) and thus lower soil CO 2 emissions (Tavares et 339 al., 2015). CO2 cumulative emissions were significantly lower in the jasmine than in the 340 paddy field, consistently with the fact that the soil carbon concentration was higher in the 341 paddy than the jasmine field (Wang et al., 2015a). Soil carbon is the substrate for the 342 microbial growth, and thereby its consumption increases the respiration and CO2 emissions 343 (Dias et al., 2010; Carbone et al., 2011; De Deyn et al., 2011). Moreover, the observed lower 344 rice root C:P, N:P ratios, and higher P concentration than in jasmine roots (Wang et al., 345 2015d; Wang et al., 2016b; Wang et al., 2017a), potentially linked to higher growth rates and 346 general activity, could also influence in the higher CO2 emitted from soil in rice than in 347 jasmine crops. 348 12 349 350 4.3. Effects of changing from rice to jasmine production on CH4 emissions 351 CH4 emissions were positively correlated with salinity, temperature, water content and pH in 352 paddy field whereas in jasmine soils, CH4 emissions were not correlated with these soil 353 variables. In paddy soils, the higher salinity can provide the nutrients that promote the CH4 354 production and emission. Temperature is the limited factors, when the temperature increment, 355 the soil production also increased (Wang et al., 2017b). Moreover, CH4 production needs 356 anaerobic environments (Minamikawa et al., 2014), such as those generated during water 357 flooding periods in paddy fields. 358 Soil pH is also another factor that controls CH4 production. Our study areas have a 359 relatively acid soil, whereas the maximum CH4 production occurs at neutral pH (Wang et al., 360 2017b). Despite of this, we did not observe correlations between soil pH and CH4 in both 361 studied crops. The CH4 emissions varied among the seasons but not consistently for both crop 362 fields. In paddies, CH4 emissions were lower soon after rice transplantation when the soil 363 was not strictly anaerobic, and also during the final ripening and drainage periods, specially 364 for early paddy. These results agreed with those observed by Minamikawa et al. (2014). But 365 the CH4 emissions in jasmine field did not change significantly in total (accumulative) and in 366 most sampling days. Jasmine fields are in dryland soils with scarce anaerobic conditions not 367 favourable for methane production (Minamikawa et al., 2014), although some production can 368 exist in anaerobic microsites of soils that supported methanogenesis even under non-flooded 369 conditions (Bhattacharyya et al., 2012). 370 CH4 cumulative emissions were significantly lower in the jasmine than in the paddy 371 field as a result of both the higher soil carbon concentration in the paddies than in the jasmine 372 fields (Wang et al., 2015a) and the more anaerobic condition in the paddies during flooding 373 period favoring the anaerobic environment necessary for methane production (Minamikawa et 374 al., 2014). Lowering of the water table or soil water content decreased the abundance of the 375 methanogenic archaeal population and hence CH4 production and increased the abundance of 376 methanotrophs and thus CH4 oxidation (Ma and Lu, 2011), thereby the CH4 emissions were 377 higher in paddy than jasmine field. 13 378 4.4. Effects of changing from rice to jasmine production on N2O emissions 379 N2O emissions were positively correlated with [Fe2+], [Fe3+], [total Fe], and negatively 380 correlated with salinity, and water content in paddy fields and positively correlated with 381 salinity, and temperature, and negatively correlated with pH, [Fe3+], and [total Fe] in jasmine 382 crops. The higher soil temperature might favor a greater N mineralization rate and hence N2O 383 production (Granli and Bøckman, 1994). As the pulses in NH4+ availability after fertilization 384 have been related to increment of N2O production by NH4+ oxidation (Pathak et al., 2002), 385 thereby the high water content during flooding decreased N2O production. Soil pH was 386 negatively related to N2O emissions in jasmine crops. Previous studies have consistently 387 observed that under alkaline treatment, N2O production was decreased (Wang et al., 2015c). 388 N2O emissions were also negatively related with iron soil concentrations in both crop types. 389 Soil Fe3+ concentration increment will enhance the formation of iron plaque on the crop 390 roots and thus limiting the N2O transport to the atmosphere (Butterbach-Bahl et al., 1997; 391 Wassmann and Aulak, 2000). Moreover, when soil Fe, Fe3+ concentrations increase, the rate 392 of Fe3+ reduction also increases, and reduced Fe 2+ accumulates in the soil (Wang et al., 393 2015c), which could inhibit microbial activity (Wu et al., 2012) and thus lower soil N2O 394 emissions (Tavares et al., 2015). 395 N2O cumulative emissions were higher in the jasmine than the paddy field. N2O 396 emission was low throughout the rice growing season. The paddies in this region are 397 strongly N limited (Wang et al., 2015c), so together with the low levels of soil O2, make that 398 most of the N2O produced is likely reduced to N2, leading to very low emissions or even a net 399 uptake of N2O (Zhang et al., 2010). For the jasmine field, the higher N2O emissions occurred 400 mainly in the growth period because the N fertilizer is in the form of NH4+, and when the soil 401 was dry, the NH4+ can be oxidized to N2O. Pulses in NH4+ availability after fertilization have 402 been related to increments of N2O production (Pathak et al., 2002). Moreover, the lower [Fe3+] 403 in the jasmine field was also associated with these higher N2O emissions. Fewer Fe3+ cations 404 in the jasmine field would decrease the competition with nitrate as electron acceptors, thereby 405 increasing the production of N2O. These fewer Fe3+ cations together with the lower pH in the 406 jasmine field could account for the indirect effects generating higher N2O emissions. 14 407 408 4.5. Integrated analysis and best management practices to reduce GWP 409 In general, CO2 and CH4 emissions were both lower five years after a paddy field was 410 converted to jasmine production, but by different mechanisms. The conversion directly 411 decreased soil CO2 emissions and increased CH4 emissions but indirectly increased CO2 412 emission and decreased CH4 emissions through its effects on soil properties. The total soil 413 emissions of both gases finally decreased. The conversion from rice to jasmine production 414 increased N2O emissions due to decreases in soil pH and [Fe3+]. These results strongly 415 suggest that the different management (fertilization), the flood/dry periods in the paddy field 416 versus the more constant soil water content in the jasmine field and probably the different 417 plant structure (grass versus tall shrub) greatly affected overall GHG emissions. The trend to 418 substitute rice with jasmine production in some areas of China should thus contribute to 419 eliminating the increases (or slight decreases) in GHG emissions in China. 420 CO2 was the most important GHG for the GWPs, and the total GWP was significantly 421 lower in the jasmine than the paddy field. Changes in soil properties associated to the changes 422 in land uses play a relevant role in the decrease of GWPs in this cropland China area. A 423 greater control of changes in soil pH and an improvement of the loads of Fe3+ during 424 fertilization are advisable to further decrease GHG emissions by the conversion from rice to 425 jasmine production. 426 427 5. Conclusions and implications 428 1. CO2 and CH4 emissions were lower, N2O increased, but total GWP of the emissions 429 decreased in the Jasmine crop with respect to the rice crop. 430 2. Soil properties also varied significantly among the seasons. Soil salinity, water content, pH, 431 [Fe2+], [Fe3+] and [total Fe] were significantly lower in the jasmine than the paddy field, but 432 temperature was significantly higher in the jasmine field. The changes in soil properties in the 433 conversion from rice to jasmine production had larger effects on the GHG emissions. 434 3. The conversion from rice to jasmine production decreased GWP emissions, and a greater 15 435 control of changes in soil pH and an improvement of the loads of Fe3+ during fertilization are 436 advisable to further decrease GWP emissions. 437 438 Acknowledgements 439 The authors would like to thank Hongchang Ren, Xuming Wang and Qinyang Ji for their 440 assistance with field sampling. Funding was provided by the National Science Foundation of 441 China (41571287; 31000209), Natural Science Foundation Key Programs of Fujian Province 442 (2018R1101006-1), Fujian Provincial Outstanding Young Scientists Program (2017), Spanish 443 Government grant CGL2013-48074-P, Catalan Government grant SGR 2014-274 and 444 European Research Council Synergy grant ERC-SyG-2013-610028, IMBALANCE-P. 445 446 447 References 448 Asensio, D., Peñuelas, J., Llusià, J., Ogaya, R., Filella, I., 2007. Interannual and interseasonal 449 soil CO2 efflux and VOC exchange rates in a Mediterranean holm oak forest in response 450 to experimental drought. Soil Biol. Biochem. 39, 2471-2484. 451 Atkin, O.K., Tjoelker, M. G., 2003. Thermal acclimation and the dynamic response of plant 452 respiration to temperature. Trends Plant Sci. 8, 343-351. 453 Barton, K., 2012. MuM. In: Multi-model inference. R package version 1.7.2. http://CRAN. 454 R-283 project.org/package=MuMIn. at http://cran.r-project.org/ package=MuMIn>. 455 Bhattacharyya, P., Neogi, S., Roy, K.S., Dash, P.K., Tripathi, R., Rao, K.S., 2013. Net 456 ecosystem CO2 exchange and carbon cycling in tropical lowland flooded rice ecosystem. 457 Nutr. Cycl. Agroecosys. 95, 133-144. 458 Bua, B., Ojirot, M., 2014. Assessing the Importance of Rice as Food and Income Security 459 Crop in Puti-puti Sub-county, Pallisa District, Uganda. Am. J. Exp. Agr. 4(5), 532. 460 Butterbach-Bahl, K., Papen, H., Rennenberg, H., 1997. Impact of gas transport through rice 461 cultivars on methane emission from rice paddy fields. Plant Cell Environ. 20, 1175-1183. 462 Carbone, M.S., Still, C.J., Ambrose, A.R., Dawson, T.E., Williams, A.P., Boot, C.M., 463 Schaeffer, S.M., Schimel, J.P., 2011. Seasonal and episodic moisture controls on plant and 464 microbial contributions to soil respiration. Oecologia 167, 265-278. 16 465 Dai, X., Yuan, Y., Wang, H., 2016. Changes of anaerobic to aerobic conditions but not of 466 crop type induced bulk soil microbial community variation in the initial conversion of 467 paddy soils to drained soils. Catena 147, 578-585. 468 De Deyn, G., Quirk, H., Oakley, S., Ostle, N., Bardgett, R., 2011. Rapid transfer of 469 photosynthetic carbon through the plant-soil system in differently managed species-rich 470 grasslands. Biogeosciences 8, 1131-1139. 471 Dias, A.T.C., van Ruijven, J., Berendse, F., 2010. Plant species richness regulates soil 472 respiration through changes in productivity. Oecologia 163, 805-813. 473 Emmett, B., Beier, C., Estiarte, M., Tietema, A., Kristensen, H., Williams, D., Peñuelas, J., 474 Schmidt, I, Sowerby, A., 2004. The response of soil processes to climate change: results 475 from manipulation studies of shrublands across an environmental gradient. Ecosystems 7, 476 625-637. 477 FAO [Food and Agricultural Organization of the United Nations]., 2009. OECD-FAO 478 Agricultural Outlook 2011–2030. 479 Fimmen, R.L., Jr, D.D. R., Vasudevan, D., Williams, M.A., West, L.T., 2008. Rhizogenic 480 Fe–C redox cycling: a hypothetical biogeochemical mechanism that drives crustal 481 weathering in upland soils. Biogeochemistry 87(2), 127-141. 482 Fox, J., Nie, Z., Byrnes, J., 2013. sem: Structural Equation Models. 483 Gaddanakeri, S.A., Bellakki, M.A., Biradar, D.P., Balikai, R.A., Nadagouda, S.V.S.V., 484 Srinivas, S., 2008. Effect of flood on pH, EC, nutrient and insect pest status in 485 sugarcane fields, in Bhima and Krishna river belts of northern Karnataka. Int. J.Agr. Stat. 486 Sci. 4(1), 157-161. 487 Granli, T., Bøckman, O.C., 1994. Nitrous oxide from agriculture. Norw. J. Agr. Sci. Suppl 12, 488 1–128. 489 Gribsholt, B., Kostka, J. E., Kristensen, E.. 2003. Impact of fiddler crabs and plant roots on 490 sediment biogeochemistry in Georgia saltmarsh. Mar. Ecol. Progress Series259, 491 237-251. 492 17 493 Gudzic, N., Aksic, M., Djikic, A., Gudzic, S., Knezevic, J., Jelić, M., 2015. Soil fertility 494 changes trends and available Fe, Mn and Zn in a long-term system user nitrogen 495 fertilizers. Int. J. Agri. Agri. R. 7, 14-20. 496 Houghton, R.A., Hackler, J.L., Lawrence, K.T., 1999. The US carbon budget: contributions 497 from land-use change. Science 285(5427), 574-578. 498 Hu, Z., Wu, S., Ji, C., Zou, J., Zhou, Q., Liu, S., 2016. A comparison of methane emissions 499 following rice paddies conversion to crab-fish farming wetlands in southeast China. 500 Environ. Sci. Pollut. R. 23(2), 1505-1515. 501 Huang, S., Rui, W., Peng, X., Liu, W., Zhang, W., 2009. Responses of soil organic carbon 502 content and fractions to land-use conversion from paddy field to upland. Environ. Sci. 30, 503 1146-1151. 504 Inglett, K.S., Inglett, P.W., Reddy, K.R., Osborne, T.Z., 2012. Temperature sensitivity of 505 greenhouse gas production in wetland soils of different vegetation. Biogeochemistry 506 108, 77-90. 507 Li, X., Xia, J., Tian, W.. 2017. Research on the water stability and the driving forces of soil 508 aggregate after paddy field switched to tea garden. J.Soil Water Conser. 31, 148-153. 509 Lin, X.B., Yan, Y.Y., Min, Q.W., Wang, J.F., Zheng, J.M., Zhang, Y.X., Wang, W.Q., 2014. 510 Evaluating the non-use value of agricultural heritage across Jasmine planted and tea 511 cultural systems in Fuzhou. Resour. Sci. 36, 1089-1097. 512 Linn, D.M., Doran, J.W., 1984. Effect of water-filled pore space on carbon dioxide and 513 nitrous oxide production in tilled and nontilled soils. Soil Sci. Soc. Am. J. 48, 514 1267-1272. 515 Liu, Y., Yang, M., Wu, Y., Wang, H., Chen, Y., Wu, W., 2011. Reducing CH4 and CO2 516 emissions from waterlogged paddy soil with biochar. J. Soil. Sediment. 11, 930-939. 517 Lu, R.K., 1999. Analytical methods of soil agrochemistry. China Agricultural Science and 518 Technology Press, Beijing. 519 Ma, K., Lu, Y., 2011. Regulation of microbial methane production and oxidation by 520 intermittent drainage in rice field soil. FEMS Microbial Ecol. 75, 446-456. 521 Minamikawa, K., Fumoto, T., Itoh, M., Hayano, M., Sudo, S., Yagi, K., 2014. Potential of 522 prolonged midseason drainage for reducing methane emission from rice paddies in 18 523 Japan: a long-term simulation using the DNDC-Rice model. Biol. Fert. Soils 50, 524 879-889. 525 Myhre, G., Shindell, D., Bréon, F.M., Collins, W., Fuglestvedt, J., Huang, J., Koch, D., 526 Lamarque, J.F., Lee, D., Mendoza, B., Nakajima, T., Robock, A., Stephens, G., Takemura, 527 T., Zhang, H., 2013. Anthropogenic and Natural Radiative Forcing. In: Climate Change 528 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth 529 Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., 530 Qin, D., Plattner, G.K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., 531 Midgley, P.M. (eds.)]. Cambridge University Press, Cambridge, United Kingdom and 532 New York, NY, USA. pp 714. 533 Pathak, H., Bhatia, A., Prasad, S., Singh, S., Kumar, S., Jain, M.C., Kumar, U., 2002. 534 Emission of nitrous oxide from rice-wheat systems of Indo-Gangetic Plains of India. 535 Environ. Monit. Assess. 77, 163-178. 536 Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., Core, T.R., 2016. nlme: Linear and Nonlinear 537 Mixed Effects Models. R package version 3.1-126, http://CRAN. 538 Ren, H.C., Lin, X.B., Wang, C., Wang, W.Q., Min, Q.W., Zhang, Y.X., Zheng, J.M., 2015. A 539 study of the residents' cognition and attitudes towards the protection of agricultural 540 heritage system from the place identity perspective: a case study of Fuzhou Jasmine and 541 Tea Cultural System. Acta Ecol. Sin. 35, 6808-6813. 542 Sharma, S.K., Shayam, D., 2012. Effect of paddy straw and paper mill effluent on 543 physico-chemical properties of soil. An Asian J. Soil Sci. 8, 12-18. 544 Slot, M., Wright, S.J., Kitajima, K., 2013. Foliar respiration and its temperature sensitivity 545 in trees and lianas: in situ measurements in the upper canopy of a tropical forest. Tree 546 Physiol. 33, 505-515. 547 Tavares, R.L.M., Farhate, C.V.V., de Souza, Z.M., Junior, N.L.S., Torres, J.L.R., Campos, 548 M.C.C., 2015. Emission of CO 2 and soil microbial activity in sugarcane management 549 systems. Afr. J. Agr. Res. 10, 975-982. 550 Treat, C.C., Wollheim, W.M., Varner, R.K., Grandy, A.S., Talbot, J., Frolking, S., 2014. 551 Temperature and peat type control CO 2 and CH4 production in Alaskan permafrost 552 peats. Global Change Biol. 20, 2674-2686. 19 553 Vogel, J.G., Bronson, D., Gower, S.T., Schuur, E.A.G., 2014. The response of root and 554 microbial respiration to the experimental warming of a boreal black spruce forest. Can. 555 J. Forest Res. 44, 986-993. 556 Wang, H., Guan, D., Zhang, R., Chen, Y., Hu, Y., Xiao, L., 2014. Soil aggregates and organic 557 carbon affected by the land use change from rice paddy to vegetable field. Ecol. Eng. 70, 558 206-211. 559 Wang, W., Wang, C., Sardans, J., Min, Q., Zeng, C., Tong, C., Peñuelas, J., 2015a. 560 Agricultural land use decouples soil nutrient cycles in a subtropical riparian wetland in 561 China. Catena 133, 171-178. 562 Wang, W., Sardans, J., Lai, D.Y.F., Wang, C., Zeng, C., Tong, C., Liang, Y., Peñuelas, J., 563 2015b. Effects of steel slag application on greenhouse gas emissions and crop yield over 564 multiple growing seasons in a subtropical paddy field in China. Field Crop. Res. 171, 565 146-156. 566 Wang, W., Lai, Y.F.D., Sardans, J., Wang, C., Datta, A., Pan, T., Zeng, C., Bartrons, M., 567 Penuelas, J., 2015c. Rice straw incorporation affects global warming potential differently 568 in early vs. late cropping seasons in Southeastern China. Field Crop. Res. 181, 42-51. 569 Wang, W. Q., Wang, C., Sardans, J., Zeng, C. S., Tong, C., Peñuelas, J., 2015d. Plant invasive 570 success associated with higher n-use efficiency and stoichiometric shifts in the 571 soil-plant system in the minjiang river tidal estuarine wetlands of china. Wetl. Ecol. 572 Manag. 23, 1-16. 573 Wang, W., Zeng, C., Sardans, J., Wang, C., Zeng, D., Peñuelas, J., 2016a. Amendment with 574 industrial and agricultural wastes reduces surface-water nutrient loss and storage of 575 dissolved greenhouse gases in a subtropical paddy field. Agr. Ecosyst. Environ. 231, 576 296-303. 577 Wang, W., Min, Q., Sardans, J., Wang, C., Asensio, D., Bartrons, M., Peñuelas, J., 2016b. 578 Organic cultivation of jasmine and tea increases carbon sequestration by changing plant 579 and soil stoichiometry. Agro. J. 108, 1636-1648. 580 Wang, W., Sardans, J., Wang, C., Zeng, C., Tong, C., Bartrons, M., Peñuelas, J., 2017a. 581 Steel slag amendment increases nutrient availability and rice yield in a subtropical paddy 582 field in China. Exp. Agr. doi:10.1017/S0014479717000412. 583 Wang, W., Sardans, J., Wang, C., Zeng, C., Tong, C., Asensio, D., Peñuelas, J., 2017b. 584 Relationships between the potential production of the greenhouse gases CO2, CH4 and 20 585 N2O and soil concentrations of C, N and P across 26 paddy fields in southeastern 586 china. Atmos. Environ. 164, 458-467. 587 Wang, W., Neogi, S., Lai, D. Y. F., Zeng, C., Wang, C., Zeng, D., 2017c. Effects of industrial 588 and agricultural waste amendment on soil greenhouse gas production in a paddy field in 589 southeastern china. Atmos. Environ. 164, 239-249. 590 Wassmann, R., Aulakh, M.S., 2000. The role of rice plants in regulating mechanisms of 591 methane emissions. Biol. Fert. Soils 31, 20-29. 592 Wu, P., Zhang, G., Li, J., Lu, H., Zhao, W., 2012. Effect of Fe2+ concentration on biomass 593 accumulation and energy metabolism in photosynthetic bacteria wastewater treatment. 594 Bioresource Technol. 119, 55-59. 595 Wu, H., Zhang, J., Chen, G., Wang, Q., Wang, L., Zhang, Y., 2016. Study on plant diversity 596 of rural herbs under different land uses in northern Zhejiang province. Guangxi Plant 36, 597 824-831. 598 Xu, C.T., 2012. Industry research report of Fuzhou Jasmine Tea. Chin. Hortic. Abstr. 28, 599 24-27, 12. 600 Xu, G.Q., Fang, S.H., Xia, T., 2001. Advances of Jasmine Tea processing technology. Tea 601 Fujian 23, 10-12. 602 Yang, D., Zhang, M., 2014. Effects of land-use conversion from paddy field to orchard farm 603 on soil microbial genetic diversity and community structure. Eur. J. Soil Biol. 64, 604 30-39. 605 Yang, J.F., Fu, T.L., Ye, N.X., Chen, Q., Zheng, N.H., Sun, Y., Yuan, D.S., Guo, Y.Q., Chen, 606 Y.R., Lin, G.S., Su, B.C., Yang, W., 2008. Fujian Jasmine tea. Xiamen: Xiamen 607 University Press, 55-57. 608 Zhang, A., Cui, L., Pan, G., Li, L., Hussain, Q., Zhang, X., Zheng, J., Crowley, D., 2010. 609 Effect of biochar amendment on yield and methane and nitrous oxide emissions from a 610 rice paddy from Tai Lake plain, China. Agri. Ecosyst. Environ. 139, 469-475. 611 Zhu, 2006. System of rice intensification. Beijing: China Agricultural Science and 612 Technology Press. 613 21 614 Table 1 615 Results of mixed models (lme function in nlme package in R, and with GLMM of MuMIn package 616 function to estimate the variance explained by fixed variables and total model) with different gas 617 emissions and studied soil variables as dependent factors, crop type (rice and jasmine) as independent 618 fixed factor and month of measurement nested within plot as random factor. 619 Variable (dependent) Model and statistics CO2 emission lme(CO2 ~ field type, random=~1|plot/month) Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, R2c = variance explained by fixed + random variables) F=5.03 R2m=0.05 P=0.031 R2c=0.34 CH4 emission lme(CH4 ~ field type, data=dades, random=~1|plot/month) Fixed variable statistics Model R2 (R2m =variance explained by fixed variable, R2c = variance explained by fixed + random variables) F=11.7 R2m=0.14 P=0.0016 R2c=0.15 N2O emission lme(N2O ~ field type, data=dades, random=~1|plot/month) Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, R2c = variance explained by fixed + random variables) F=7.8 R2m=0.095 P=0.0084 R2c=0.13 Soil salinity lme(soilsal ~ field type, random=~1|plot/month) Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, R2c = variance explained by fixed + random variables) F=165 R2m=0.52 P<0.0001 R2n=0.78 Soil temperature lme(T ~ field type, random=~1|plot/month) Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, 22 R2n = variance explained by fixed + random variables) F=1.98 R2m=0.0048 P=0.17 R2c=0.83 Soil water content lme(water ~ field type, random=~1|plot/month Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, R2c = variance explained by fixed + random variables) F=134 R2m=0.65 P<0.0001 R2n=0.66 Soil pH lme(soilpH ~ field type, random=~1|plot/month) Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, R2n = variance explained by fixed + random variables) F= 735 R2m=0.91 P<0.0001 R2n=0.92 Soil [Fe2+] lme(Fe2 ~ field type, random=~1|plot/month) Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, R2c = variance explained by fixed + random variables) F=104 R2m = 0.54 P<0.0001 R2c = 0.63 Soil [Fe3+] lme(Fe3 ~ field type, random=~1|plot/month) Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, R2c = variance explained by fixed + random variables) F= 74.0 R2m = 0.35 P<0.0001 R2c = 0.66 Soil [total Fe] lme(totalFe ~ field type, random=~1|plot/month) Fixed variable statistics Model R2 (R2m = variance explained by fixed variable, R2c = variance explained by fixed + random variables) F=91.4 R2m=0.43 P<0.0001 R2c=0.67 620 23 Variable General model lme(Variable ~ field type, data=dades, random=~1|plot/month) Fixed variable Model R2 (R2m = variance explained by statistics fixed variable, R2n = variance explained by fixed + random variables) CO2 emission F=5.03 R2m=0.05 P=0.031 R2c=0.34 CH4 emission F=11.7 R2m=0.14 P=0.0016 R2c=0.15 N2O emission F=7.8 R2m=0.095 P=0.0084 R2c=0.13 Soil salinity F=165 R2m=0.52 P<0.0001 R2n=0.78 Soil temperature F=1.98 R2m=0.0048 P=0.17 R2c=0.83 Soil water content F=134 R2m=0.65 P<0.0001 R2n=0.66 Soil pH F= 735 R2m=0.91 P<0.0001 R2n=0.92 Soil pH F= 735 R2m=0.91 P<0.0001 R2n=0.92 Soil [Fe2+] F=104 R2m = 0.54 P<0.0001 R2c = 0.63 Soil [Fe3+] F= 74.0 R2m = 0.35 P<0.0001 R2c = 0.66 Soil [total Fe] F=91.4 R2m=0.43 P<0.0001 R2c=0.67 621 622 24 623 Table 2 624 Global-warming potential (GWP) of paddy and jasmine fields (average±SE, N=3) analyzed by mixed models, with plot as a random factor. Land use Cumulative greenhouse-gas emission (g m-2) GWP (kg CO2-eq ha-1) Total GWP CO2 CH4 N2O CO2 CH4 N2O (kg CO2-eq ha-1) Paddy field 7245±335a 25.5±1.3a 0.56±0.08a 72453±3352.48a 8865±428.13a 1667±2284a 82784±3421a Jasmine field 4744±125b 0.31±0.44b 3.36±1.15a 47437±1245.69b 105±150b 10013±3423b 57556±3056b 625 Different letters within a column indicate significant differences (P<0.05). 626 627 628 629 630 631 632 633 634 635 25 636 Table 3 Summary of the RM-ANOVAs for the greenhouse-gas emissions and soil properties for land use. Greenhouse gases df MS F P CO2 Types 1, 4 1467851.493 48.926 0.002 Time 11, 44 1684375.311 93.170 <0.001 Types × Time 11, 44 899110.027 49.733 <0.001 CH4 Types 1, 4 148.652 356.113 <0.001 Time 11, 44 39.305 49.797 <0.001 Types × Time 11, 44 38.700 49.031 <0.001 N2O Types 1, 4 1840123.191 5.921 0.072 Time 11, 44 544852.929 6.144 <0.001 Types × Time 11, 44 545865.122 6.155 <0.001 Soil properties Soil salinity Types 1, 4 1.765 942.596 <0.001 Time 11, 44 0.106 39.766 <0.001 Types × Time 11, 44 0.028 10.592 <0.001 Soil temperature Types 1, 4 11.871 15.016 0.018 Time 11, 44 199.616 386.741 <0.001 Types × Time 11, 44 17.988 34.850 <0.001 Soil water content Types 1, 4 12358.837 498.721 <0.001 Time 11, 44 263.692 24.063 <0.001 Types × Time 11, 44 270.197 24.657 <0.001 Soil pH Types 1, 4 41.283 1168.044 <0.001 Time 11, 44 0.173 17.536 <0.001 Types × Time 11, 44 0.144 14.670 <0.001 2+ Soil Fe Types 1, 4 205.010 690.008 <0.001 Time 11, 44 8.863 71.891 <0.001 Types × Time 11, 44 5.997 48.642 <0.001 3+ Soil Fe Types 1, 4 1012.485 1333.573 <0.001 Time 11, 44 118.453 88.080 <0.001 Types × Time 11, 44 41.953 31.196 <0.001 Soil total Fe Types 1, 4 2128.854 4219.880 <0.001 Time 11, 44 178.700 120.264 <0.001 Types × Time 11, 44 72.470 48.772 <0.001 26 637 Table 4 638 Average soil salinity, temperature, water content, pH, [Fe2+], [Fe3+] and [total Fe] (average±SE, N=3) analyzed by mixed models, with plot as a random factor. Land use Salinity (mS cm-1) Temperature (°C) Water content (%) pH [Fe2+] (mg [Fe3+] (mg [Total Fe] (mg g-1) g-1) g-1) Paddy field 0.443±0.032a 21.0±1.14b 55.8±2.06a 6.35±0.03a 4.47±0.35a 15.4±1.07a 19.9±1.39a Jasmine field 0.130±0.016b 21.8±0.79a 29.6±0.94b 4.84±0.03b 1.10±0.11b 7.88±0.56b 8.97±0.56b 639 Different letters within a column indicate significant differences (P<0.05). 640 641 642 643 644 645 646 647 648 27 649 Table 5 650 Pearson correlation coefficients between soil traits and influencing factors. Land use Gas Salinity Temperature Water content pH [Fe2+] [Fe3+] [Total Fe] Paddy field CO2 0.28* 0.58** 0.35* -0.29* -0.58** -0.59** -0.60** (N=36) CH4 0.65** 0.55** 0.58** -0.07 -0.28* -0.48** -0.44** N2O -0.36* -0.26 -0.34* -0.14 0.28* 0.29* 0.29* Jasmine field CO2 0.73** 0.76** 0.29* -0.66** -0.16 -0.61** -0.64** (N=36) CH4 0.01 0.17 0.06 -0.08 0.04 0.05 0.06 N2O 0.51** 0.55** 0.27 -0.44** -0.05 -0.48** -0.49** 651 * P<0.05, ** P<0.01. 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 28 667 Figure captions 668 Fig. 1. Location of the paddy and jasmine fields. 669 Fig. 2. Seasonal variation of CO2 (A), CH4 (B) and N2O (C) emission in the paddy and jasmine fields. 670 Different letters indicate significant differences between land uses (P<0.05). 671 Fig. 3. Soil salinity (A), temperature (B), water content (C), pH (D), [Fe2+] (E), [Fe3+] (F) and [total Fe] 672 (G) in the paddy and jasmine fields. Different letters indicate significant differences between land uses 673 (P<0.05). 674 Fig. 4. Diagrams of the structural equation models that best explained the maximum variance of the 675 CO2, CH4 and N2O emissions, with the two crop types (rice and jasmine) and soil traits as exogenous 676 factors and CO2 (A), CH4 (B) and N2O emissions as endogenous variables. Positive and negative 677 relationships are indicated by black and red arrows, respectively. The values on the arrows are 678 standard estimates with their corresponding levels of significance. The total variance explained for 679 each endogenous variable by the model (R2) is also indicated beside the corresponding variable. 680 Fig. 5. Total, direct and indirect estimates of the effects of the variables on the final endogenous 681 variable of each model: CO2 (A), CH4 (B) and N2O (C) emissions for the structural equation models 682 (SEMs) in Figure 4, with the corresponding level of significance of the standard estimates. * P<0.05, 683 ** P<0.001, *** P<0.0001. 684 685 686 687 688 689 690 691 692 693 694 695 696 29 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 Fig. 1. 717 718 719 720 721 722 723 724 725 726 727 728 30 729 730 731 3000 A a Paddy field Jasmine field a CO2 flux (mg m-2 h-1) 2000 a a b a a 1000 a a b b b a b b b a b a a a a a b 0 732 Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. 21 B a 14 CH4 flux (mg m-2 h-1) a 7 a a a b b b aa b b a a a a a a a a a a a a 0 -7 Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. 733 Sampling date (Month) 3000 C a N2O flux (μg m-2 h-1) 2000 a 1000 a a a b a a a a a a b a a a b a a b a aa 0 a Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. 734 Sampling date (Month) 735 736 Fig. 2. 737 738 739 31 740 741 1.2 36 A B Paddy field Jasmine field a a a a b a b a a a a Soil salinity (mS cm-1) 0.8 Soil temperatuew (℃) a 24 b b a b a a b a a a a b a b a a b a a 0.4 a 12 b b a a a b b b b b b a b b b b 0.0 0 742 Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. 90 9 C D a a a a a a a a a a a a a a a a a a Soil water content (%) 60 6 b a a b b b b b a b b b b b b Soil pH a a b a a b b b a b b b 30 b 3 b b 0 0 743 Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. 9 36 E a F a a Soil Fe3+concentration (mg g-1) Soil Fe2+concentration (mg g-1) a a 24 a a a 6 a a a a a a b a b a b b 3 a 12 a a a b b a a a a b b b b b b b b b b b b b b b b b b 0 0 Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Sampling date (Month) 744 36 a G a a a Soil total Fe concentration (mg g-1) a 24 a a b a a a a a b b 12 b b b b b b b b b 0 Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Sampling date (Month) 745 746 747 Fig. 3. 748 32 749 750 751 752 Crop type (A) P<0.0001 -1.91 Soil pH Soil [Fe2+] R2=0.91 R2=0.55 CO2 emission R2=0.39 Crop type (B) R2=0.52 R2=0.55 P<0.0001 Soil water 0.78 Soil salinity content Soil [Fe2+] R2=0.66 CH4 emission R2=0.49 Crop type (C) R2=0.52 R2=0.55 P=0.37 -0.33 Soil salinity Soil pH Soil [Fe3+] R2=0.91 N2O emission R2=0.30 753 754 Fig.4. 755 756 757 33 Total effects (estimates) Direct effects (estimates) Indirect effects (estimates) (A) Soil [Fe2+] *** *** Soil pH ** ** *** *** Crop type ** -2 -1 0 1 2 -1 0 1 2 -1 0 1 2 Estimated effects on CO2 emissions Total effects (estimates) Direct effects (estimates) Indirect effects (estimates) Soil water (B) content ** ** Soil [Fe2+] ** ** Soil salinity ** ** *** Crop type *** *** -1.2 -0.8 -0.4 0 0.4 0.8 1.2 -1.2 -0.8 -0.4 0 0.4 0.8 1.2 -1.2 -0.8 -0.4 0 0.4 0.8 1.2 Estimated effects on CH4 emissions Total effects (estimates) Direct effects (estimates) Indeirect effects (estimates) Soil pH *** *** (C) Soil [Fe2+] * * Soil salinity * * Crop type ** * ** -1.2 -0.8 -0.4 0 0.4 0.8 1.2 -1.2 -0.8 -0.4 0 0.4 0.8 1.2 -1.2 -0.8 -0.4 0 0.4 0.8 1.2 Estimated effects on N2O emissions 758 759 Fig. 5. 34