Rice Cultivation and Greenhouse Gas Fluxes: Monitoring, Quantification, Adaptation and Mitigation

Rice is grown in different environments ranging from tropical to temperate regions with varying climatic, edaphic, and biological conditions and agricultural management. Tropical and subtropical lowland rainfed rice is usually grown in bunded fields that are flooded with rainwater for at least part of the cropping season. On the other hand, medium upland or upland rice productions are practiced by managed irrigation applications or may be somewhat rainfed. Large amounts of water are generally used in rice production and a significant portion of the total water requirement is used for land preparation which is unique. Rice cultivation acts as an anthropogenic manipulator of ecosystem carbon (C) and nitrogen (N) dynamics through uptake, fixation, emission and transfer of C and N between biospheric, lithospheric and atmospheric pools. Rice cultivation has a potential role in the global budget of potent greenhouse gases (GHGs) such as methane (CH 4 ), carbon dioxide (CO 2 ), and nitrous oxide (N 2 O). But large variability exists in CO 2 , CH 4 and N 2 O fluxes in both spatio-temporal scales and their relative contribution changes with management practices. Therefore, field level studies for seasonal, annual and inter-annual scales are indeed needed for precise quantification of GHGs exchanges employing state-of-the art sampling and measurement technology. Hence, from the environmental sustainability point of view, the agricultural management practices for rice production should so improvised by adopting mitigation cum adaptation approaches that these could help in abatement of GHGs emission to atmosphere.


Introduction
Rice is the major food crop in Asia. About 80% of it is grown under flooded conditions [1]. In 2010, approximately 154 million ha were harvested worldwide, of which 137 million ha (88 percent of the global rice harvested) were in Asia, of which 48 million ha (31 percent of the global rice harvested) were harvested in Southeast Asia alone [2]. Rice is grown in different environments ranging from tropical to temperate regions with varying climatic, edaphic and biological conditions and under different agricultural management practices which naturally affect the rates of potent agriculturally important GHGs namely CH 4 , CO 2 and N 2 O emissions fluxes [3,4]. Carbon dioxide is the most important anthropogenic GHG. During rice cultivation CO 2 originates from burning of fossil fuels due to operation of farm machinery while soil tillage for land preparation, microbial decomposition of soil organic carbon and matter (SOC and SOM) and soil-crop respiration [5,6]. Soil CO 2 efflux is usually measured with the help of soil respiration chamber or sampling is done by manual closed / automatic chamber from soil-crop system and quantified using gas chromatography [7,8]. The state of the art eddy covariance (EC) technique is also employed for quantifying ecosystem level net ecosystem CO 2 exchange (NEE) [5].
On the other hand, flooding of irrigated rice fields produces anaerobic soil conditions which are conducive to the production of CH 4 due to the action of methanogenic bacteria [4,9]. Methane emission from rice cultivation is usually measured by manual closed / automatic chamber method using gas chromatography or by micrometeorological method like EC technique which measures net ecosystem methane exchange (NEME) [3][4][5]. Nitrous oxide is generated by the microbial transformation of N in rice soils and manures. It is often found to be enhanced where available N exceeds plant requirements, especially under wet conditions as found in rice cultivation [10]. The rice paddies act as sources of major N 2 O emission upon nitrogenous fertilizer (e.g. urea) application. Nitrous oxide is usually measured by manual closed / automatic chamber method using gas chromatography [7,8] or by micrometeorological method (EC technique), but its application is limited due to very expensive trace gas analyzer and the technologies involved [11]. Many of the factors controlling exchanges of GHGs fluxes between rice paddies and the atmosphere are different from any other agricultural practices and ecosystems because rice is flooded during most of its cultivation period [12].
Moreover, GHGs fluxes in terms of gaseous C and N between rice fields and the atmosphere are controlled by several biological and physical processes. The GHGs flux dynamics during rice cultivation follows complex pathways and shows variability at different time scales [5,13]. Therefore, field-cum-laboratory-based studies to measure gaseous C (CO 2 +CH 4 ) and N (N 2 O) fluxes from soilcrop system and to improve the understanding of the associated factors mitigating the GHGs fluxes are important. The continuous monitoring and measurement provide a useful understanding for examining the roles of different rice and rice based production systems contributing to GHGs fluxes under different agro-climatic zones and management practices [14]. The study could also be employed to explore better understanding of GHGs exchanges for scaling up gaseous C and N fluxes from point scale to and it could further be upscaled for predicting future anticipated climate change [15]. An array of factors like range of soil, climatic variables, physicochemical-biological interactions in crop-soil and agricultural management practices influence the production and subsequent emission of gaseous C and N from rice and rice-based production systems [16,17]. Hence, by manipulating soil temperature, soil moisture, pH, SOM, organic manure and fertilizer applications, water management, cultivar selection, tillage practices etc. (in totality crop-soil management and agronomic practices), either singly or in optimized combination, and by adopting judicious land use management practices, GHGs emission could be mitigated.
Sampling, monitoring techniques and estimation of greenhouse gas fluxes from rice and rice-based production systems CO 2 , CH 4 and N 2 O flux monitoring and quantification from crop-soil system using manual closed / automatic chambers: Greenhouse gas (viz. CO 2 , CH 4 and N 2 O) fluxes from the soil-crop system of rice fields are monitored by manual closed chamber method or automatic chamber method. Sampling for GHGs flux measurements are done from replicated plots in the morning (9:00-9:30h) and afternoon (15:00-15:30h) and the average of the morning and afternoon fluxes are used as the flux for the day. The gas samplings are performed soon after sowing / transplanting of the rice at close intervals throughout the cropping season. This may also be applied on bare soil in fallow periods or in rice nurseries. The sampling days generally encompass beginning from early growth stage of the crop through vegetative, reproductive phase till physiological maturity of the crop. For sampling and estimation of GHGs fluxes, generally six rice hills are covered with a Perspex chamber (53cm×37cm×51cm; length×width × height from seedling to tillering stage and 53cm×37cm×71cm; length×width×height from maximum tillering to maturity stage) and placed on aluminum base plate inserted into the soil (  A battery operated air circulation pump with an air displacement of ~1.5L min -1 and connected to polyethylene tubing is used to mix air inside the chamber and draws the air samples into gas sampling bags (tedlar bags) at fixed intervals of 0, 15 and 30min. Gas samples from the sampling bags are analyzed for CO 2 , N 2 O and CH 4 concentrations to determine respective fluxes employing gas chromatograph fitted with thermal conductivity detector (TCD) (or in sometimes using methanizer along with flame ionization detector, FID), electron capture detector (ECD) and FID, respectively. Cumulative GHGs fluxes and emission for the entire cropping period are computed by plotting the flux values against the days of sampling and is expressed in unit kg ha -1 . These chamber Copyright © Suvadip Neogi MCDA.000620. 5(4).2020 measurements are widely used as they are easy to apply in multiple small plots in fields. The automatic chamber measurements allow continuous and frequent measurements of GHGs fluxes and are expected to produce more reliable results compared to manual chamber measurements as diurnal variations of GHGs fluxes are duly taken care of. But, the automatic chamber measurements suffer from underestimation of fluxes due to chamber effects on soil moisture conditions during rainfall or pressure-temperature variations, which are corrected afterwards during post-processing of data [20,21].
Use of soil respiration chamber: Soil CO 2 fluxes are analyzed by environmental gas monitor chamber equipped with infrared gas analyzer and attached to a data logger. A flag or marker is placed at the exact spot in the plot from where a CO 2 flux is measured throughout the study period. The soil respiration chamber is placed on the soil surface for 2min in each spot until CO 2 flux measurements are recorded in data logger ( Figure 2). The CO 2 flux is recorded from the inter-row position of the rice plants. All measurements are recorded at 9:00-12:00h and 15:00-17:00h [6]. The average of the morning and afternoon fluxes are considered as daily flux and cumulative CO 2 emissions are expressed as kg ha -1 .

Use of chamber connected to automatic CH 4 analyzer:
For application of standard direct method of CH 4 flux analysis, a portable chamber is used to estimate the methane emissions from the paddy fields of the study site ( Figure 3). The chamber is made with Perspex with an exhaust fan fitted at one of the sides. The fan operates for about 5min just before sampling, in order to ensure homogeneous mixing of gases within the chamber. The suction pipe of the methane analyser is fitted into the ceiling of the hood. The Perspex chamber is used on a channel inserted into the soil prior to transplantation. During the analysis, the channel is filled with water to make it leak tight. The analysis is done on field using a methane analyser equipped with FID [22,23].  disturbance to the vegetation over which fluxes are measured. The EC technique is based on high frequency (10-20Hz) measurements of wind speed and wind direction as well as CO 2 and CH 4 concentrations at a point over the canopy using a three-axis sonic anemometer and a fast response infrared gas analyzer. The two sensors, three-axis sonic anemometer and fast response infrared gas analyzer/methane analyzer, are normally installed at 1.5-3m height (depends on the rice canopy). The mean vertical flux density of CO 2 or CH 4 is obtained as the 30minute covariance between vertical fluctuations of air density and the CO 2 or CH 4 mixing ratio. A positive covariance indicates net GHG (CO 2 or CH 4 ) transfer into atmosphere and a negative value indicates net GHG absorption by the vegetation (CO 2 ) / soil (CH 4 ). The NEE is further partitioned into gross primary production (GPP) and ecosystem respiration (RE). An eddy covariance system integrated with trace gas analyzer (TGA), a tunable diode laser analyzer, can measure trace gas fluxes viz. N 2 O. Eddy covariance measurements of N 2 O using lead salt tunable diode laser (TDL) spectrometers and quantum cascade laser (QCL) spectrometers are being also practiced, but due to high cost involved, its application is somewhat limited [11].

Factors for GHGs fluxes and emission quantifications from rice and rice-based production systems
The atmospheric concentration of CO 2 has drastically increased from 280 parts per million by volume (ppmv) in 1750 (preindustrial revolution) to 402ppmv in present day [25] and is currently increasing at the rate of 1.9ppmv yr -1 [26]. Atmospheric CH 4 concentration has increased from about 715 to 1827 parts per billion by volume (ppbv) at present [27] over the same period and is increasing at the rate of 7ppbv yr -1 [26]. Similarly, the atmospheric concentration of N 2 O has increased from about 270ppbv in 1750 to 325ppbv at present [27] and is increasing at the rate of 0.8ppbv yr -1 [26]. In terms of global warming potential (GWP) amongst these three GHGs for a period of 100-year time horizon the order is like N 2 O> CH 4 > CO 2 [26]. Agriculture accounted for an estimated emission of 5.1 to 6.1Gt CO 2 -eq year -1 , around 10-12% of total global anthropogenic emissions of GHGs, where CH 4 contributed 3.3Gt CO 2 -eq year -1 and N 2 O 2.8Gt CO 2 -eq year -1 [26]. Of global anthropogenic GHG emissions, agriculture accounted for about 60% of N 2 O and 50% of CH 4 [26].

Factors for CO 2 fluxes and emission quantifications
There are several factors influencing CO 2 production and emission from the soil. These include soil texture, moisture, pH and salinity which influence CO 2 production through their effect on soil microbial activity and root respiration. Apart from these, external factors viz. tillage, irrigation, temperature, fertilizer and manure application also have an effect on CO 2 production and emission [28]. During rice cultivation tillage practices, SOM decompositions and soil-crop respirations (heterotrophic + autotrophic) are mainly responsible for CO 2 emission to the atmosphere. But in case of lowland flooded rice ecology the total CO 2 emissions due to respirations are often found to be nullified by the gross primary production (photosynthesis) and the resultant net CO 2 flux remains negative (negative NEE) representing the ecosystem as net CO 2 -C sink [5,29]. Different scenario is experienced in upland aerobic rice production [30].
During crop season (heading to maturity stage; chemical fertilization and drained soil condition), daily NEE (CO 2 flux) in rice in Taiwan varied from -17.0 to 12.9μmol CO 2 m -2 s -1 .
The NEE was positive during night, average value ~ 2.8μmol CO 2 m -2 s -1 ; whereas during the daytime the flux was negative, average value ~ -1.2μmol CO 2 m -2 s -1 [9]. The NEE exhibited a clear diurnal pattern varying between -38 to 10μmol CO 2 m -2 s -1 during full heading stage of rice in Bangladesh (chemical fertilization and flooded soil condition) [31]. In rice fields in Japan, NEE varied between -16.9 to 9.3μmol CO 2 m -2 s -1 (chemical fertilization and drained soil condition; 1 month before heading stage) and -19.6 to 4.3μmol CO 2 m -2 s -1 (chemical fertilization and flooded soil condition; 1 month before heading stage) due to different management practices [32]. Seasonal integrated NEE varied from -258g Cm -2 (flooded rice fields) to -85g Cm -2 (aerobic rice fields) in IRRI,
Annual average soil respiration rates and total soil respiration of paddy soil in the subtropical region of China were estimated to be 178.5-259.9mg m -2 h -1 and 1.56-2.28kg m -2 yr -1 , respectively. The soil respiration rates at night times during the fallow periods (in absence of rice crops) were 52-398mg m -2 h -1 [33]. Introduction of minimum tillage (MT) in rice-based production systems led to significantly reduce CO 2 emission from soil. The CO 2 -C emissions, as a whole, were 24% higher in between plants than in rows, and were in the range of 23. The CO 2 -C emission was found highest under maize (44%) followed by rice (33%) and cowpea (23%) irrespective of CT and MT practices [6]. Combined application of rice straw (RS) and green manuring (GM) showed higher seasonal CO 2 -C flux (~1858kg ha -1 ) than control (~1100kg ha -1 ) in lowland submerged rice. But due to higher grain yield, the emission per unit yield was significantly lower [8].

Factors for CH 4 fluxes and emission quantifications
Methane is released to the atmosphere by ebullition, diffusion across floodwater-air interface and by transport through aerenchyma [3, 4,7,8]. In undisturbed paddy fields ~90% of CH 4 emission occurs through aerenchyma [34]. Flooded rice soils are favorable for the production and emission of CH 4 due to methanogenic bacteria which utilize readily decomposable organic compounds under anaerobic soil condition [3,4]. Both CH 4 production and emission from flooded rice soils are strongly driven by several soil processes namely changes in soil redox status and pH, dynamics of substrate and nutrient availability etc. [35]. Besides, CH 4 emission is also influenced by cultivation practices and agricultural management viz. chemical fertilizer and agrochemicals application, organic matter amendment etc.
The net emission from an agricultural system is the result of production (methanogenesis) and consumption (methanotrophy). Whether the net emission is positive or negative depends on the relative magnitudes of these processes. However, CH 4 production in rice fields depends on soil characteristics (organic carbon, OC), rice varieties (especially morphology and physiology), management practices (fertilizer application, water management, land preparations). Continuous flooding, pure mineral fertilizer, organic manure amendments, root exudates, and cultivar types have found to influence on CH 4 emission [36]. Methane emissions exhibit large spatio-temporal variations. Among the sources of CH 4 , irrigated rice fields are estimated to contribute between 6-8% of the total 410-660 million tons yr -1 emitted globally [9]. Methane emission varied from 14 to 375mg m -2 d -1 in most rice growing areas in the world [37].
Annual global estimation of CH 4 emission from flooded rice fields was accounted for 7.08Tg [38]. In Thailand CH 4 emission was estimated ~99kg CH 4 ha -1 season -1 from deep water rice fields [39].
Average CH 4 emission rates ranged between 11-364mg m -2 d -1 in rice fields of Beijing, China [40]. In India the mean CH 4 emission from rice fields ranged between 3.5-4.2Tg yr -1 [41]. Bhatia et al. [42] estimated 4.7Tg yr -1 CH 4 emission from the Indian paddy fields with the highest emission of 1.379Tg yr -1 from the irrigated rice fields. For the first time in India, employing EC system in lowland submerged rice ecology Bhattacharyya et al. [5] quantified NEME ~84 and 129kg ha -1 in two consecutive rice growing seasons and the annual NEME was ~262kg ha -1 . Alberto et al. [43] estimated seasonal CH 4 emission in irrigated rice fields of IRRI and it has been found that the ecosystem released 3.03g Cm −2 of CH 4 -C to the atmosphere during the growing period.

Factors for N 2 O fluxes and emission quantifications
Nitrous oxide is a byproduct of denitrification and nitrification processes in soil. Nitrification is the main source of N 2 O production followed by emission under aerobic conditions, while denitrification dominates under flooded rice fields [7,8]. Nitrification is the process of oxidation of ammonium form of N to nitrite or nitrate form and subsequent emission of N 2 O from soil. On the other hand, denitrification is the microbial reduction of nitrate or nitrite form of N to dinitrogen or oxides of N under anaerobic condition [3,4]. Several microbiological and ecological factors influence N transformation processes in soil and hence N 2 O production-emission. Nitrous oxide is primarily emitted in pulses after fertilization and strong rainfall events. Land use practices and N fertilizer applications greatly influence N 2 O emission from soil.
Type and dose of nitrogenous fertilizer controls the amount of N flows through the system and hence influences the N 2 O emission [44]. Increase in total N 2 O emission with the increase in N application rates is one of the responsible factors [45]. In general, soil water content, water filled pore spaces, relative abundance of electron donors (soil organic carbon) and acceptors (primarily oxygen, nitrate, and sulfate), degradable organic matter, soil texture, pH and salinity regulate N 2 O emission from soil [46]. Chao et al. [47] estimated around 0.67Mg N 2 O-N yr -1 from the paddy fields of Taiwan. Nitrous oxide emission from the Chinese rice fields ranged between 39-164mg N m -2 h -1 [48]. Sharma et al. [49] estimated N 2 O emissions from irrigated and upland paddy fields of India ~4-210 and 2-10Gg yr -1 , respectively. Bronson et al. [50] observed N 2 O fluxes in an irrigated rice system were generally negligible during the growing seasons, but small peaks (~3.5mg N 2 O-N m -2 d -1 ) appeared after N fertilizer applications.
The N 2 O flux increased sharply during the drainage period. Higher N 2 O flux (~80mg N 2 O-N m -2 d -1 ) found during fallow period was due to nitrification of mineralized organic N in the topsoil and possibly from denitrification in the wet subsoil [51]. Kumar et al. [45] observed total N 2 O-N emissions during crop growth season in an irrigated rice system ranged ~235g N 2 O-N ha -1 with application of ammonium sulphate and ~160g N 2 O-N ha -1 with urea application. Worldwide at differently managed rice ecologies (in terms of ecology, irrigation applications and varying N fertilizer doses) broadly the total seasonal N 2 O emission varied from ~0.15 to ~0.61g m -2 [52,53].

Mitigation cum adaptation strategies for reducing gaseous C and N emissions from rice and rice-based production systems
Mitigation cum adaptation strategies for reducing CO 2 emissions from rice and rice-based production systems: Carbon dioxide emission depends on the soil-water management, cultivation methods and agricultural operations [54]. Overlying floodwater and its prolonged existence of on rice soil creates anaerobic environment. This helps in slow decomposition of SOM. Moreover, it acts as a diffusion barrier to liberated CO 2 from soil to the atmosphere, thereby storing more C in soil [3][4][5]29,30,43]. That's why the aerobic rice cultivation system in upland condition is responsible for higher CO 2 emissions as compared to lowland flooded rice soils. No tillage or minimum tillage practices reduce CO 2 efflux from soil to the atmosphere rather than conventional tillage and foster more C sequestration in soil [6]. Mulching effect on soil surfaces reduces CO 2 emissions from soils, retains soil moisture and, thereby, helps to sequester more C in soils [55]. Overall, Soil management practices like increasing SOC content, reduced / conservation tillage, manuring, residue incorporation, improving soil biodiversity, micro-aggregation and mulching can play an important role in sequestering C in soil [28].

Mitigation cum adaptation strategies for reducing CH 4 emissions from rice and rice-based production systems:
Rice cultivation is often regarded as culprits for CH 4 production and emission [56]. Mitigation cum adaptation strategies for reducing CH4 emissions from rice and rice-based production systems: (i) judicious water management [57], (ii) use of situation specific rice cultivars [58], (iii) efficient fertilizers management [58], (iv) manipulation of cropping practices [59], (v) Effective land management [60] etc.
Overall, these options mainly target management of the crop, soil and irrigation requirements, varietal choice and agrochemical usage [3,4,7,8,61,62]. The higher age of seedlings at the time of transplanting showed significant impact on subsequent CH 4 emission [63]. Hence, the proper selection of rice cultivars is a potential adaptation strategy for reduced CH 4 emission. Experimentally screened rice cultivars with proven low CH 4 emission potential in a specific situation / ecology may be selected. Methane emission rates were found higher in transplanted rice than direct sown rice [64]. Mid-season drainage has been found to substantially reduce CH 4 emissions by about 30-50% as compared to continuous flooding maintained in rice fields during crop cultivation [64]. Intermittent irrigation applications or cycles of alternate wetting and drying (AWD) as occur in rainfed rice situations led to significant reductions in the CH 4 emissions from rice fields [65].
Land management in the winter crop season has been found to significantly influence CH 4 emission fluxes following flooded rice growing period [66]. Land management practices in the winter crop season also affected temporal variation patterns of CH 4 fluxes and soil redox potential (Eh) after flooding. Therefore, water management in the preceding crop season is a crucial factor in influencing CH 4 emission from rice fields [67]. The application of sulphate fertilizers is a suitable option to reduce CH 4 emissions by increasing alternative electron acceptors in soil. The partial competition of the sulphate-reducing bacteria with methanogens for C substrate plays crucial role in soil [68]. Application of single super phosphate and potassium fertilizer also led to the decreased cumulative seasonal CH 4 emissions.
Sulphur contained in the single super phosphate decreased the CH 4 emissions. Potassium helps in maintaining higher levels of oxidation status in the topsoil profile and encourages oxidation processes in the rhizosphere zone [69]. The influence of crop physiology (rhizodeposition) on seasonal CH 4 emissions is important factor. Reduction of the rates of rhizodeposition, therefore, is beneficial to both yield and reduced CH 4 emissions [70]. Rice straw (RS) composted properly resulted in a six-fold reduction in CH 4 emission compared with partially composted RS [71]. Dual cropping of Azolla in conjunction with urea considerably reduced CH 4 efflux without affecting the rice yields and can be used as a practical adaptation option for minimizing CH 4 flux from flooded paddy [72].

Mitigation cum adaptation strategies for reducing N 2 O emissions from rice and rice-based production systems:
Direct relation exists between nitrogen use efficiency (NUE) and N 2 O emissions [73]. These strategies for N 2 O emission reduction include matching N supply with demand, use of slow release fertilizers, urease and nitrification inhibitors, proper forms of fertilizer, appropriate rate, dose and method of application etc. [74]. Application of N in splits as per the crop requirement is an important strategy to improve NUE, minimization of N loss and checking N 2 O emission from soil-crop system [75]. Application of customized leaf colour chart (CLCC) depending on the variety and ecology where the crop is grown for checking leaf greenness and use of SPAD meter in this respect is helpful for determining number of splits, doses and application time as per crop requirement [76]. Addition of nitrification inhibitors viz. dicyandiamide along with urea has been instrumental in reducing N 2 O loss [45,77]. Use of neem oil and karanj oil coatings on urea has been found to reduce N 2 O emissions [78]. Drip irrigation system has been found to reduce the N 2 O emissions compared to the furrow irrigation in arid and semi-arid areas [79]. Denitrification related N 2 O losses from urea application in flooded rice ecology are reduced when urea is deep placed compared to surface broadcast [80]. Application of slow/ controlled release fertilizer (e.g., polyolefin-coated ammonium nitrate) has been found to impart 3-7 fold reduction in N 2 O emissions compared to uncoated ammonium sulfate from arable soil [81].

Conclusion and Future Road Map
The impact of GHGs on climatic conditions and the influence of such climatic change on rice productivity is now a reality, although there is a need to assess the extent of such influences. The holistic studies of GHGs emission flux quantification followed by subsequent validation and upscaling from different rice ecologies and approaches adopted towards GHGs emission mitigation are still lacking. Judicious land use and appropriate management practices (viz. adoption of conservation agriculture and introduction of low carbon resource conservation technologies) would help to mitigate the process of GHGs emission. The GHGs emission depends on the soil and water management, cultivation methods and agronomic operations. The options for GHGs emission reduction largely depend upon the situation and component specific factors.
Thus, the optimized mitigation approaches would be so formulated to include management of the crop, soil and irrigation requirements, physico-chemical-biological factors, varietal choice and agrochemicals usage. Hence, the C and N biogeochemistry in agricultural ecosystems (crop-soil system) consisting of the holistic interactions of soil temperature, moisture, pH, redox potential, substrate concentration profile, decompositions and transformations driven by ecological drivers viz. climate, crop / vegetation and management practices need to be studied meticulously for formulation-implementation of better mitigation cum adaptation strategies keeping in mind better soil healthquality and sustained crop productivity.