Using 13C isotopic analysis to assess soil carbon pools associated with tile drainage management during drier and wetter growing seasons

Agricultural Water Management 192 (2017) 232–243

Contents lists available at ScienceDirect

Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat

Research Paper

Using 13 C isotopic analysis to assess soil carbon pools associated with tile drainage management during drier and wetter growing seasons Alisha Van Zandvoort a,∗ , Ian D. Clark a , Corey Flemming a , Emilia Craiovan b , Mark D. Sunohara b , Natalie Gottschall b , Ronda Boutz c , David R. Lapen b,∗ a

Advanced Research Complex, University of Ottawa, 25 Templeton Street, Ottawa, Ontario, K1N 6N5, Canada Ottawa Research and Development Centre, Agriculture and Agri-Food Canada, 960 Carling Avenue, Ottawa, Ontario, K1A 0C6, Canada c South Nation Conservation, 38 Victoria Street, P.O. Box 29, Finch, Ontario, K0C 1K0, Canada b

a r t i c l e

i n f o

Article history: Received 10 March 2016 Received in revised form 12 May 2017 Accepted 19 June 2017 Keywords: Controlled tile drainage Soil respiration Gas diffusion Carbon-13 Greenhouse gas emission Agricultural fields

a b s t r a c t Controlled tile drainage (CTD) is an agricultural management practice with well-documented water quality and agronomic benefits. However, due to field water drainage abatement effects, the practice could potentially impact soil respiration. The ␦13 C of soil gas samples were analyzed from silt loam CTD and uncontrolled tile drained (UTD) fields cropped with corn, soybean, and forage to test whether CTD affects total soil respiration (RT ) through altering residue decomposition and/or gas transport. We also examined whether CTD imparted an over tile (OT) and between tile (BT) locational effect on RT . All gas samples are from bulk rhizosphere and microbial sources; except for the forage crop measurements which included plants growing inside the sampling chamber collars. The mean values of ␦13 C for RT were not statistically different (p > 0.05) between four of the five CTD and UTD field pairs. Neither were the means of ␦13 C of RT different (p > 0.05) between OT and BT locations in four of the five CTD fields. The generalized lack of statistical differences in ␦13 C of RT observed, may have resulted from the drier and wetter seasonal conditions muting drainage management effects on surface soil water contents. The results could be used to help refine the carbon footprint of tile drainage management practices. Crown Copyright © 2017 Published by Elsevier B.V. All rights reserved.

1. Introduction Soil respiration is a significant source of atmospheric CO2 (IPCC, 2013; Rochette et al., 1999a; Smith et al., 2007). Soil respiration originates from plants, through rhizosphere respiration, and soils, through microbial respiration (Cheng, 1996; Rochette et al., 1999a; Ryan and Law, 2005). The rate that CO2 is produced within the soil is influenced by nutrient availability, soil temperature, and soil water content (Kreba et al., 2013; Phillips et al., 2010; Schwen et al., 2015). Further, gaseous diffusion, a primary transport mechanism associated with gas exchange in the soil environment, is strongly influenced by soil physical properties (Ball et al., 1997; Jin and Jury, 1996; Marshall, 1959; Moldrup et al., 2004, 2013; Tuli et al., 2005). Agricultural crop lands cover a significant amount of the Earth’s land surface (Smith et al., 2007), making them an important contributor to atmospheric CO2 . In light of climate change implications, it is becoming increasingly important to understand and

∗ Corresponding authors. E-mail addresses: [email protected] (A. Van Zandvoort), [email protected] (D.R. Lapen). http://dx.doi.org/10.1016/j.agwat.2017.06.012 0378-3774/Crown Copyright © 2017 Published by Elsevier B.V. All rights reserved.

quantify the carbon footprint of agricultural management practices (Paustian et al., 2000; West and Marland, 2002). Agricultural management practices can impart strong controls over soil CO2 emissions (Al-Kaisi and Yin 2005; Lal, 2004; Mordhorst et al., 2014; Paustian et al., 1997, 2000). For example, no-till or reduced tillage practices have been shown to reduce soil respiration rates compared to intensive or conventional tillage practices (Al-Kaisi and Yin 2005; Fortin et al., 1996; Paustian et al., 2000; Sánchez et al., 2002; Schwen et al., 2015). Controlled tile drainage (CTD), an agricultural management practice that regulates tile drain discharge (Gilliam et al., 1979), has well-documented water quality and agronomic benefits (Mejia et al., 2000; Ng et al., 2002; Sunohara et al., 2015; Wesstrom and Messing, 2007). However by virtue of its effect upon soil and groundwater hydrology, CTD could potentially augment greenhouse gas emissions to the atmosphere (Elmi et al., 2000; Nangia et al., 2013). For example, Nangia et al. (2013) found some significantly higher bulk CO2 emissions from soil surfaces in cropped CTD fields in relation to paired uncontrolled tile drained (UTD) fields; but that study did not reveal the source and nature of the respired carbon or whether gas diffusion was directly affected by CTD.

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

We are currently unaware of studies that have analyzed the effects of agricultural water table management, such as CTD, on the origin of soil respiration and soil gas diffusivity. A useful tool in this regard is stable 13 C, which can be used as a natural tracer to analyze alterations in gas transport and the dynamics of soil organic carbon; including identification of carbon substrates being oxidized (Cerling et al., 1991; Clark, 2015; Drewitt et al., 2009; Gregorich et al., 1995; Phillips et al., 2010; Rochette et al., 1999a). By measuring ␦13 C of total soil respiration (RT ; combined respiration from roots of plants, rhizosphere, soil organisms) from cropped agricultural fields, we can evaluate contributions from both soil and plant on total respiration, and the associated influences of agricultural management practices like CTD. Using ␦13 C as a natural tracer is made possible by the differing ␦13 C values of C3 and C4 plants as a result of the C3 and C4 photosynthetic pathways (Cheng, 1996; Drewitt et al., 2009; Gregorich et al., 1995; Rochette et al., 1999b). In general, the ␦13 C values of plants are more depleted than atmospheric CO2 due to discrimination against 13 CO2 during CO2 fixation. C3 plants have a considerably more depleted ␦13 C value (∼−26 to −28‰) than atmospheric CO2 (∼−8‰) because of the inefficient step of CO2 respiration that results in close to 20‰ fractionation. In contrast, C4 photosynthesis reduces photorespiration and increases CO2 fixation, which results in a lower fractionation (∼−6‰) and a more enriched ␦13 C (∼−10 to −14‰) compared to C3 plants (O’Leary, 1988). The ␦13 C of RT can be used to identify the carbon substrate being oxidized because the ␦13 C of plants affect the ␦13 C of soil organic matter (SOM). The ␦13 C of SOM resembles the ␦13 C of the plant material from which it was derived such that soils derived from C4 plants have a more enriched ␦13 C (∼−12‰ to −14‰) compared to soils derived from C3 plants (∼−24‰ to −29‰) (Cheng, 1996; Gregorich et al., 1995). Thus, by growing a C4 crop in a soil containing a C3 isotope signature (i.e. a soil where the SOM component predominantly originated from C3 plants), or vice versa, results in isotopically distinct values for the crop and the SOM; thus providing a means to quantify the source of RT (Cheng, 1996; Drewitt et al., 2009; Gregorich et al., 1995; Rochette et al., 1999b). Identifying the source of RT is useful for assessing how field management practices sequester, emit, and cycle soil carbon; especially within the context of gauging the environmental footprint of a best management practice (BMP). The ␦13 C of RT can be used to analyze the dynamics of SOM because the ␦13 C of RT is affected by biological changes in the produced CO2 and soil physical changes that alter gas exchange with the atmosphere (Cerling et al., 1991; Phillips et al., 2010). At steady state diffusion, the ␦13 C of RT equals the ␦13 C produced within the soil from either plants or soil (Cerling et al., 1991). Changes in the environment can alter the amount of CO2 produced and the gas diffusivity through the soil thus creating non-steady state conditions where the ␦13 C of RT does not equal the ␦13 C produced within the soil (Phillips et al., 2010). Therefore, the ␦13 C of RT can identify whether soil physical changes occurred that affected CO2 production and/or gas transport. While agricultural studies have used ␦13 C of RT to reveal the sources of respired carbon (Drewitt et al., 2009; Rochette et al., 1999a, 1999b); as far as we are aware, there are no reports that have used ␦13 C of RT to reveal whether CTD affects the carbon substrates being oxidized and/or the soil physical conditions that govern CO2 production and gas transport. We hypothesize that CTD will alter residue decomposition and soil gas diffusivity through impacts on soil hydrology (e.g., decreased water table depth and increased soil water contents reducing oxidation of organic matter). We also hypothesize that the location in a field with respect to a tile drain will be important in this regard due to water table mounding that can occur between tile drains (Smedema and Rycroft, 1983), different soil chemical/physical properties over tiles (trenches backfilled

233

with disturbed soil during time of tile installation) (Nangia et al., 2013), and biopores that can preferentially occur over tile drains, in relation to between them (Nuutinen et al., 2001). As such, for two growing seasons with contrasting precipitation amounts, the overall objectives of this study were to: (1) measure and compare the ␦13 C of RT from corn, soybean, and perennial forage fields under contrasting tile drainage management treatments (i.e., controlled tile drainage (CTD) and conventional or uncontrolled tile drainage (UTD)), (2) evaluate whether ␦13 C of RT in a CTD field differs with respect to tile drain (ie., over tile versus between tile) location, and (3) evaluate relationships among ␦13 C of RT and environmental variables, such as soil temperature, soil water content, and water table depth, that can be directly impacted by CTD.

2. Materials and methods 2.1. Study fields and weather The fields studied were located within experimental watersheds in eastern Ontario, Canada (Fig. 1; see Sunohara et al. (2015) for details). The fields were low-gradient (slope < 1%) and soil at the site was classified as a Bainsville silt loam (Wicklund and Richards, 1962). For additional details on specific soil properties and hydrology see Sunohara et al. (2014). The 30-year normal total annual precipitation (1981–2010) in the area is ∼981 mm and average daily air temperature ∼6.5 ◦ C (ECCC, 2015). From June to end of September (the approximate growing season considered here), the average 30-year (1981–2010) rainfall and air temperature is 369 mm and 18 ◦ C, respectively (ECCC, 2015). During the study, total growing season rainfall was 281 mm and 460 mm in 2012 and 2013, respectively. Average air temperature during the growing season was 19.2 ◦ C (SD = ±5.7) and 17.8 ◦ C (SD = ±5.6) for 2012 and 2013, respectively. This study was conducted during the growing season of 2012 and 2013 on subsurface tile drained agricultural fields under corn (Zea mays), soybean (Glycine max), and forage/hay crop [combination of Phleum pretense (timothy grass) and Medicago sativa (alfalfa)] (Fig. 1). This study examined two pairs of fields in 2012 and three pairs of fields in 2013, with each field pair under a common cropping-soil management but varying in the type of tile drainage management practice (UTD and CTD field pairing) (Table 1). Tillage practices for soybean in 2012 (Fields 1 & 2) consisted of fall mouldboard ploughing (to max ∼0.3 m depth), and spring cultivation using a chisel plough. On field pairs 11 & 14 and 12 & 13, forage was killed off with herbicide in 2012, and no-tillage soybeans were subsequently planted in 2013. Row spacing for corn and soybean was approximately 0.5 m with typical seeding rates and standard mineral fertilizer rates (∼170 kg N/ha for corn and ∼10 kg N/ha for soybean) for the region (Ontario Ministry of Food and Rural Affairs, 2009; Sunohara et al., 2015). The crop history and the associated photosynthetic pathway for field pairs 11 & 14 and 12 & 13 for the five years prior to the study were corn (C4 : 2007–2010) and sudan grass (C4 : 2011), whereas for field pair 1 & 2 it was corn (C4 : 2007, 2009, 2010) and soybean (C3 : 2008, 2011). Tiles in the fields were ∼0.1 m diam., spaced 15–17 m apart, and had a nominal maximum field depth of roughly 1 m. For specific fields in each pair, lateral tile drains connected to a header tile (∼0.15 m in diameter), which was in turn connected to a main field-specific outlet that drained into the drainage ditch. For both CTD and UTD fields, in-line water flow control structures (Agri Drain Corp., Adair, IA) were installed on these outlets (see Sunohara et al., 2014, 2015). Water flow control was employed for CTD fields by setting structure stopgates (stoplogs) to a flow control depth of roughly 0.6 m below the surface; when water levels were shal-

234

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

Fig. 1. The location of the experimental fields in Eastern Ontario, Canada, and graphical layout of experimental set-up. Table 1 Planting, harvest, and key tile drainage management dates for field pairs studied. Field pair

CTD field: area (ha)

UTD field: area (ha)

Year

Planting date: crop type

Date tile drain control started

Date tile drain control stopped

Harvest date

1&2

Field 2:2.3

Field 1:2.0

2012 2013

May 16: soybean May 5: corn

May 16 May 29a

September 19 September 26b

September 24 November 14

11 & 14

Field 11:4.2

Field 14:4.1

2012

In 2011: forage

May 23

November 19

2013

May 15: soybean

May 27a

September 26b

June 1, July 10, August 20, October 24 October 25

May 15: soybean

a

September 26b

October 25

12 & 13 a b

Field 12:5.0

Field 13:4.2

2013

May 27

CTD flow structure stopgates were left open for a longer period of time because farmer wanted them open due to excessive rainfall. Due to late season rainfall, farmer desired stopgates to be opened (free drainage) in late September.

lower than this depth, tile flow to the receiving stream occurred. Flow control was imposed between roughly spring planting and near time of harvest in fall (Table 1). For CTD fields, tile drainage control was not imposed during the non-growing season. For UTD fields, tile drains were allowed to drain unimpeded year round (no stopgates were used to control flow in the in-line structures).

2.2. Soil-respired CO2 and isotope measurements Gas sampling was carried out using the nonsteady-state chamber method described by Rochette and Bertrand (2008) and Rochette and Hutchinson (2005), which was also similarly used by Nangia et al. (2013). Briefly, chamber bases were installed in the soil between crop rows (corn and soybean), except for forage

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

fields where forage covered the entire field and thus the chambers contained growing plants (making measurements not purely RT for forage). Vegetation was not cleared from forage chambers in order to minimize soil disturbance, and to ensure consistency with measurements made in corn and soybean fields where vegetation was not disturbed. Rectangular chamber bases (0.75 × 0.15 m) were used in soybean fields, and square chamber bases (0.35 × 0.35 m) were used in corn fields. Both types of chamber bases had flat lids. Forage fields had circular chamber bases of 0.41 m diam., and a lid of height 0.32 m to accommodate growing plants. Plastic chamber lids contained a rubber septum sampling port on the top and a foam gasket seal on their underside. Chamber bases were installed at ∼0.1 m depth in the soil (with ∼0.05 m depth of the chamber remaining out of the soil to create a headspace) and left there for the growing season. Headspace volumes were measured routinely. Gas sampling was carried out between 10:00–14:00 h at weekly intervals from over tile (OT) and between tile (BT) locations on both fields of each field pair (Fig. 1); sample numbers varied marginally due to logistics or weather. In field pair 1 & 2, gas sampling was made in one area of the field in 2013 (total of 12 chambers per field) and two areas of the field in 2012 (total of 8 chambers per field) (Fig. 1). In field pair 11 & 14, gas sampling was made in one area of the field with a total of 4 chambers per field in 2012 and 12 chambers per field in 2013 (Fig. 1). In field pair 12 & 13, gas sampling in 2013 was carried-out in one area of the field with a total of 12 chambers per field (Fig. 1). Gas sampling involved placing the chamber lid on top of the chamber base, and using a 30 mL syringe (fitted with a 26-gauge needle) to immediately collect a 30 mL gas sample from the chamber headspace at time zero. The sample was injected into 12 mL ® septum-capped glass vials (Exetainers , Labco Ltd.), with a silicone septum that contained 2–3 mg of magnesium perchlorate (Rochette and Bertrand, 2008). During gas sampling, four gas samples were collected from each chamber’s headspace and they were analyzed for CO2 concentration. Gas samples from soybean and corn fields were sampled at 0, 6, 12, and 18 min whereas forage fields were sampled at 0, 12, 24, and 36 min to account for larger chamber volumes (Rochette and Bertrand, 2008; Rochette and Hutchinson, 2005). A fifth gas sample from each chamber was collected from soybean and corn fields at 24 min and from forage fields at 37 min, and these were analyzed for ␦13 C. In addition, two samples were collected in 2013 in the study area in order to determine the atmospheric concentration of CO2 (i.e. [CO2 ]atm ). This [CO2 ]atm was used to correct the ␦13 C values of the gas samples. The collected gas samples were analyzed for CO2 concentration using the Varian CP-3800 gas chromatograph with CombiPal autosampler with helium as the carrier gas. The injection volume was 5 mL and a flame ionization detector was used. Samples of seven reference standard gases (containing CO2 concentrations of: 0, 362, 1015, 2049, 5142, 10106, 19812 ppmv) were handled the same way as the collected gas samples and approximately one was inserted for every five gas samples during the analysis. CO2 fluxes were calculated using the rate of concentration change inside the chamber during deployment as per Rochette and Bertrand (2008), and Rochette and Hutchinson (2005). The collected gas samples were analyzed for ␦13 C using a continuous flow gasbench + DeltaPlusXP IRMS at the G.G. Hatch Stable Isotope Lab, University of Ottawa, Ontario, Canada. Isotopic compositions are expressed as permil difference (‰) referenced to the International standard VPDB (1). ␦13 C = [(13 C/12 Csample )/(13 C/12 Cstandard )]-1

␦13 Cemitted = [(␦13 Cmeasured × [CO2 ]measured )

(1)

235

Fig. 2. Total daily rainfall and soil temperature (0–0.15 m depth) in 2012 and 2013, from the weather station in Field 1. Dates are dd/mm/yy.

− (␦13 Catm × [CO2 ]atm )]/[CO2 ]emitted

(2)

[CO2 ]measured = (slopeCO2-time × timeCO2-sample ) + [CO2 ]atm

(3)

[CO2 ]emitted = [CO2 ]measured − [CO2 ]atm

(4)

␦13 C

The of the gas samples were corrected for [CO2 ]atm by removing the contribution of atmospheric CO2 present inside the chamber (2). ␦13 Cmeasured was the sample’s result obtained from the IRMS, and [CO2 ]measured was the calculated concentration of CO2 in the sample at the time of sampling. [CO2 ]measured was calculated using (3) where, slopeCO2-time is the slope (ppmv/min) obtained by plotting the CO2 concentrations against sampling time for each chamber-based measurement. TimeCO2-sample represented the elapsed time at sample collection (i.e., 24 min for soybean and corn fields, and 37 min for forage fields). The closest location to our site that contained ␦13 Catm data was Fraserdale, Ontario (linear distance of 700 km to study site); the ␦13 Catm for the approximate growing season (May to end of September) was −8.54 ± 0.70‰ (for

236

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

400

Table 2 Average ± standard deviation for water table depth below surface (m) for field pairs studied. p values are presented for t-tests. Note: measurement time frames may differ between some field pairs.

2012 2013 30-year normal

Field Pair

UTD

CTD

p value

300

2012 Field 1 & 2 Field 11 & 14

1.57 ± 0.42 2.69 ± 0.63

1.59 ± 0.41 2.49 ± 0.67

0.000 0.000

200

2013 Field 1 & 2 Field 11 & 14 Field 12 & 13

1.12± 0.30 1.28± 0.38 1.41 ± 0.32

0.9 7± 0.41 1.07 ± 0.45 1.2 7± 0.42

0.000 0.000 0.000

100

Se p

Au g

Ju l

Ju n

M ay

0

Fig. 3. Cumulative rainfall for the 2012 and 2013 growing seasons from the weather station at Field 1. The average 30-year total rainfall (1981–2010) for (June to end of September) is presented as reference (ECCC, 2015).

2009 and 2010) (Huang, 2015). Thus, ␦13 Catm was set at −8‰ for the current study because the difference between −8‰ and −8.54‰ was insignificant for the resulting calculations. [CO2 ]atm was set at 400 ppm because the average of the two air samples collected at the site in 2013 was 393.3 ± 3.6 ppm and the value was subsequently rounded to 400 ppm. The CO2 emitted from the soil was calculated using (4).

2.3. Isotopic analysis of soils and plants Soil samples (0–0.15 m, 0.15–0.30 m, and 0.30–0.60 m depth) were collected once a month from all fields with a hand auger (0.076 m length and 0.047 m ID cores) and analyzed for ␦13 C of SOM. Soil sampling for field pair 1 and 2 in 2012 and 2013 was carried out at three roughly equidistant locations down the middle of each field from both OT and BT locations. Soil sampling for field pairs 11 & 14 (in 2012 and 2013) and 12 & 13 (in 2013) was conducted from two OT and two BT locations. SOM was prepared for ␦13 C analysis using the method described by Brodie et al. (2011). Briefly, the soil was dried, ground, placed in silver capsules and acidified with HCl to remove carbonates. These prepared soil samples were analyzed for ␦13 C on the IRMS at the G.G. Hatch Stable Isotope Lab, University of Ottawa. Just prior to harvest, plant samples were collected from OT and BT locations from all fields and were separated into their respective plant parts as indicated below. Soybean plant samples, for 2012, were separated into beans and remainder of the plant, as compared to partitioning out pods (containing the beans), leaves, roots, and stems from the soybean plants in 2013. Corn was separated into cobs, leaves, seeds, roots, and stems in 2013. Forage plant samples included the whole plant sans the roots. For field pairs 1 & 2 (2012 and 2013), 11 & 14 (2013), and 12 & 13 (2013), composite plant samples for each field were analyzed for ␦13 C by compositing 30 plants, 4 plants, and 6 plants, respectively. In field pair 11 & 14 in 2012, forage samples were collected from a destructive sampling area of 0.38 m2 , where samples analyzed for ␦13 C from 1st, 2nd, 3rd, and 4th forage cuts during the growing season were a composite of plants collected in 3, 2, 2, and 3 destructive sampling areas respectively. In addition, from the 4th forage cutting, one forage sample from OT and BT sites (each a composite of 2 plants) was analyzed for each field in the pair.

The composite plant (part) samples were dried at 80 ◦ C, and finely ground (<100 mesh). Two grams from each plant (part) sample were mixed for composite samples based on location over or between tile lines (i.e., OT and BT), and all plant samples were analyzed for ␦13 C on the IRMS at the G.G. Hatch Stable Isotope Lab, University of Ottawa. 2.4. Measurements of environmental variables Water table monitoring sites (Fig. 1) consisted of 3 m deep groundwater wells composed of perforated PVC pipe (0.05 m ID) wrapped in geotextile cloth. Wells were located midway between tile drain laterals. Each groundwater well contained a level logger that consisted of a pressure transducer water level sensor and datalogger combination (WL16 Water Level Loggers, Global Water Instruments, Inc., Gold River, CA). The data was used to characterize water table depth below surface (WTDBS) at 15 min intervals throughout the growing season. Volumetric soil water content (0–0.20 m depth) (SWC20), soil temperature (0–0.10 m depth) (ST10), and air temperature were measured within 0.5 m of each gas sampling chamber at time of sampling. SWC20 of the soil was measured using the ESI (Environmental Sensors Inc., Moisture Point, Model MP-917, Victoria, BC) time domain reflectometry system. ST10 and air temperature were measured using a Fluke 52-2 thermocouple thermometer and temperature probe (Fluke Corporation, WA). Two weather stations (HOBO Weather Station Data Logger, Onset Computer Corp., MA), located in Field 14 and Field 1 measured precipitation, soil temperature (0–0.15 m depth) (ST15), and air temperature at 30 min intervals throughout the growing season. 2.5. Statistical analysis Two sample t-tests (Minitab 17) were used to determine if the means of ␦13 C of RT among CTD and UTD fields were different on a field pair basis. Also, two sample t-tests were used to compare mean differences for environmental variables (ST10, SWC20, and WTDBS) among CTD and UTD fields in the same way. In addition, t-tests were used to compare means of OT and BT ␦13 C of RT within CTD fields. 3. Results and discussion 3.1. Weather conditions and environmental variables Average growing season air temperature was similar during both growing seasons (Fig. 2), but total rainfall differed; with 460 mm and 281 mm during the 2013 and 2012 growing seasons, respectively. There was 24% less and 25% greater precipitation in 2012 and 2013, respectively, compared to the average 30-year rainfall (1981–2010) for the approximate growing season (June to end of September = 369 mm) (ECCC, 2015) (Fig. 3). The average ST15 for

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

237

Table 3 Average ± standard deviation for volumetric soil water content (0–0.20 m depth), soil temperature (0–0.10 m depth), and daily water table depth below surface at gas sampling times during the 2012 and 2013 growing seasons. p values are presented for t-tests. Note: measurement time frames may differ between some field pairs. Year

Field pair

Soil temperature (◦ C)

Soil water content (% vol.)

Daily water table depth below surface (m)

CTD

UTD

p value

CTD

UTD

p value

CTD

UTD

p value

2012

1&2 11 & 14

20.49 ± 8.04 22.19 ± 8.55

20.65 ± 8.67 24.05 ± 8.14

0.942 0.444

20.30 ± 3.36 17.60 ± 5.78

20.03 ± 3.11 17.15 ± 5.17

0.739 0.779

1.61 ± 0.46 2.65 ± 0.59

1.62 ± 0.44 2.84 ± 0.51

0.923 0.157

2013

1&2 11 & 14 12 & 13

32.20 ± 4.05 33.80 ± 4.51 31.57 ± 3.85

31.78 ± 4.50 33.34 ± 5.16 31.64 ± 4.12

0.326 0.552 0.918

18.33 ± 4.00 19.34 ± 3.62 19.32 ± 2.90

19.05 ± 4.76 19.41 ± 3.52 19.36 ± 2.82

0.242 0.912 0.928

1.10 ± 0.31 1.20 ± 0.32 1.48 ± 0.19

1.21 ± 0.18 1.40 ± 0.19 1.57 ± 0.11

0.002 0.000 0.001

Table 4 Average ± standard deviation for volumetric soil water content (0–0.20 m depth) and soil temperature (0–0.10 m depth) measured at gas sampling times for over tile drain and between tile drain measurements. p values are presented for t-tests. Note: measurement time frames may differ between some field pairs. Soil temperature (◦ C)

Soil water content (% vol.) Year and crop

CTD field

Over tile

Between tile

p value

Over tile

Between tile

p value

2012 Soybean Forage

2 11

20.51 ± 8.27 24.12 ± 7.51

22.33 ± 8.96 24.69 ± 6.72

0.467 0.805

20.76 ± 3.56 18.01 ± 6.35

20.81 ± 3.48 17.61 ± 6.03

0.961 0.876

2013 Corn Soybean Soybean

2 11 12

32.05 ± 3.32 33.41 ± 4.24 31.47 ± 3.62

32.39 ± 4.65 33.33 ± 4.18 31.67 ± 4.08

0.558 0.902 0.764

18.34 ± 4.07 19.39 ± 3.86 18.54 ± 2.35

18.39 ± 4.02 19.40 ± 3.80 19.15 ± 2.73

0.951 0.989 0.340

Table 5 Monthly average ± standard deviation for ␦13 C (‰) of soil organic matter for all studied fields throughout the 2012 and 2013 growing seasons. The soil is composite samples from 0 to 0.60 m depth. Note: Fields 11–14 in 2013 were sampled on October 1st not in September. Month

June July August September Total

Field 1

Field 2

Field 11

Field 14

Field 12

Field 13

2012

2013

2012

2013

2012

2013

2012

2013

2013

2013

−22.42 ± 1.32 −22.36 ± 0.71 −22.08 ± 0.53 −22.14 ± 0.66 −22.32 ± 1.04

−22.65 ± 0.36 −23.20 ± 0.72 −22.03 ± 0.41 −21.78 ± 0.80 −22.41 ± 0.80

−22.68 ± 1.30 −22.16 ± 0.92 −21.47 ± 0.95 −22.44 ± 1.00 −22.34 ± 1.20

−21.39 ± 0.70 −23.32 ± 1.36 −21.37 ± 0.49 −21.30 ± 0.90 −21.84 ± 1.22

−25.32 ± 0.41 −24.75 ± 0.26 −24.97 ± 0.37 −25.11 ± 0.44 −25.09 ± 0.42

−24.60 ± 0.40 −24.36 ± 0.30 −24.47 ± 0.52 −25.15 ± 0.37 −24.65 ± 0.49

−25.14 ± 0.25 −24.55 ± 0.14 −24.80 ± 0.42 −23.92 ± 1.27 −24.71 ± 0.74

−24.09 ± 0.15 −25.08 ± 0.85 −24.64 ± 0.28 −24.52 ± 0.34 −24.58 ± 0.58

−25.33 ± 0.30 −24.64 ± 0.59 −24.96 ± 0.28 −24.61 ± 0.67 −24.88 ± 0.55

−24.47 ± 0.79 −25.12 ± 0.27 −25.11 ± 0.28 −24.52 ± 0.18 −24.81 ± 0.53

the growing season (± SD), from the weather station in Field 1, was similar in 2012 (17.8 ◦ C ± 2.6) and 2013 (17.4 ◦ C ± 2.1). The WTDBS throughout the growing seasons for all CTD and UTD field pairs is presented in Fig. 4. There was a significant difference (p < 0.05), based on automated measures for the entire growing season, in WTDBS between each CTD and UTD field pair for each growing season (Table 2). In the 2013 growing season, CTD Fields 2, 11, and 12 had a shallower average WTDBS compared to their UTD field pair counterparts. However, based on gas sampling day values, there was no significant difference in the daily average WTDBS between CTD and UTD field pairs in 2012, whereas in the wetter 2013 growing season, there was a significant difference (p < 0.05), with CTD fields having shallower WTDBS (Table 3). Soil water contents for CTD and UTD field pairs are presented in Fig. 5. Both SWC20 and ST10 were not significantly different (p > 0.05) between CTD and UTD field pairs for both years of study (Table 3). Similarly, means of SWC20 and ST10 were not significantly different (p > 0.05) among OT and BT locations in all CTD fields during gas sampling times (Table 4). It was expected that a shallower WTDBS in CTD fields would result in a greater SWC20 compared to UTD fields, but this was not observed. A few possible explanations exist as to why SWC20 did not significantly vary among CTD and UTD field pairs. First, it is hypothesized that there could have been greater root water uptake by crops in the CTD fields (Cicek et al., 2010), which may have reduced differences in SWC20 among CTD and UTD fields. Shallower water table depths have been shown to increase transpiration rates and root water uptake by vegetation (Askri et al., 2014; De Silva et al., 2008; Li et al., 2015). Ng et al. (2002) found 50% greater transpiration rates from corn in a CTD treatment, which had a shallower

water table depth relative to corn in an UTD treatment. Furthermore, Mejia et al. (2000) suggested that higher water tables in CTD fields allowed greater water uptake by crops resulting in greater yields in CTD plots compared to UTD plots. On the same fields as those in this study, Kross et al. (2015) found no significant difference in crop biomass among the CTD and UTD field pairs in 2012 and 2013. This lack of difference in biomass could be taken to indicate that there may not have been strongly expressed differences in water uptake by the CTD and UTD crops. Two other viable hypotheses are proposed. In 2012, the average water table depths were greater than 1.6 m during the growing season (see Table 2); a depth that is below the level for effective tile drainage management. As such, surface soil water content differences among CTD and UTD fields would not have been expected to be strongly expressed. In the wetter 2013 growing season, there was 25% greater precipitation compared to the average 30-year rainfall (1981–2010) for the approximate growing season (ECCC, 2015), and likely the precipitation was frequent enough that surface soil moisture differences between CTD and UTD field pairs were muted. 3.2. Soil organic matter and plant ı13 C The average ␦13 C of SOM over both years of study (± SD) was −22.25 ± 1.09‰ for both Field 1 and 2, and −24.79 ± 0.58‰ for Fields 11–14. Average ␦13 C of SOM remained relatively constant temporally in each field throughout the growing season indicating that the plants did not affect the ␦13 C of SOM (Table 5). As the ␦13 C of SOM resembles the ␦13 C of the plant material from which it was derived, the SOM of Field 1 and 2 had greater carbon inputs from previous C4 plants compared to Fields 11–14, which are in the range

238

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

Fig. 4. Water table depth below surface (WTDBS) during the 2012 and 2013 growing seasons. Dates are dd/mm/yy.

Table 6 Average ± standard deviation for ␦13 C (‰) of soybean plants from Field 1 and 2 prior to harvest in 2012, and from Fields 11–14 prior to harvest in 2013. 2012

2013

Part of Soybean Plant

Field 1

Field 2

Field 11

Field 12

Field 13

Field 14

Plant sans the pods Pods Leaves Root Stem Total

−27.77 ± 0.06 −27.66 ± 0.04 ND ND ND −27.71 ± 0.07

−27.06 ± 0.52 −27.09 ± 0.40 ND ND ND −27.07 ± 0.38

ND −27.43 ± 0.05 −28.64 ± 0.37 −27.53 ± 0.05 −27.81 ± 0.12 −27.85 ± 0.53

ND −27.65 ± 0.32 −28.78 ± 0.16 −27.40 ± 0.34 −27.40 ± 0.03 −27.81 ± 0.64

ND −28.21 ± 0.66 −28.96 ± 0.17 −27.67 ± 0.18 −27.07 ± 0.12 −27.98 ± 0.79

ND −28.10 ± 0.90 −28.87 ± 0.01 −27.54 ± 0.52 −27.45 ± 0.32 −27.99 ± 0.73

ND = not determined.

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

239

Fig. 5. Mean soil water contents (0–0.20 m depth) during gas sampling days. Error bars indicate standard error. Dates are dd/mm/yy.

for SOM from C3 plants (−24‰ to −29‰) (Cheng, 1996; Gregorich et al., 1995). The average ␦13 C of corn, soybean, and forage plants for all fields and years of study were −12.37 ± 0.89‰, −27.80 ± 0.64‰, and −29.09 ± 0.97‰ respectively. These ␦13 C values were expected since the results for corn were in the common range for C4 plants (∼−11 to −15‰) whereas forage and soybean were in the common range for C3 plants (∼−26 to −30‰) (O’Leary, 1988). ␦13 C values for each part of the plant were similar to each other; the ␦13 C values for soybean and corn plants are given in Tables 6 and 7 respectively. Rochette et al. (1999a) found the ␦13 C of corn leaves (−13.44 ± 1.34‰), stems (−12.12 ± 0.93‰), and roots (−13.65 ± 0.86‰) had similar ␦13 C values. The forage plants had an

Table 7 Average ± standard deviation for ␦13 C (‰) of five different parts of corn plants from Field 1 and 2 prior to harvest in 2013. Part of Corn Plant

Field 1

Field 2

Cob Leaves Seeds Root Stem Total

−12.06 ± 0.29 −14.07 ± 1.56 −12.28 ± 0.01 −12.91 ± 0.52 −12.92 ± 1.17 −12.84 ± 1.00

−11.28 ± 0.18 −12.25 ± 0.41 −11.73 ± 0.08 −12.19 ± 0.03 −12.05 ± 0.07 −11.90 ± 0.41

average ␦13 C of −28.87 ± 1.07‰ in Field 11 and −29.31 ± 0.86‰ in Field 14.

240

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

Fig. 6. Average ␦13 C in gas samples collected from field experiments. Error bars indicate standard deviations. Dates are in dd/mm/yy.

3.3. ı13 C of soil-respired CO2 The mean ␦13 C of the gas samples did not differ (p > 0.05) between CTD and UTD field pairs throughout the two growing seasons, except for one field pair comparison (Field 11 & 14 in 2012) (Table 8) where CTD Field 11 resulted in a more enriched ␦13 C compared to UTD Field 14 (Fig. 6). The forage fields (Field 11 and 14 in 2012) were the only fields that contained plants within the gas sampling chambers. As such, the significant difference in ␦13 C of the gas samples between CTD Field 11 and UTD Field 14 in 2012 (p < 0.05) may have been the result of different amounts of biomass/plant mixes within the chamber bases. Although speculation, a greater biomass in UTD Field 14 chambers could have led to greater con-

tribution from plant respiration resulting in a more depleted ␦13 C of the gas samples (closer to the ␦13 C value of forage (∼−29‰) compared to soil (∼−25‰)). The ␦13 C of gas samples is affected by biological changes in the produced CO2 and physical changes during gas transport (Cerling et al., 1991; Clark, 2015; Phillips et al., 2010). The outward diffusion of CO2 from the high PCO2 (partial pressure of carbon dioxide) of soils to low PCO2 of the atmosphere results in an approximate fractionation of 4.4‰ as 13 CO2 has lower diffusivity than 12 CO2 (Cerling et al., 1991; O’leary, 1988; Phillips et al., 2010). At a diffusive steady-state, the ␦13 C of emitted CO2 equals the ␦13 C produced within the soil (Cerling et al., 1991). However, under non-steady state conditions caused by environmental changes such as changes

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

241

Fig. 7. Average ␦13 C in gas samples taken from chambers installed directly over tile (OT) and midway between tile laterals (BT). Error bars indicate standard deviations. Note that forage in 2012 has no error bars as only one sample was measured for each OT and BT location. Dates are in dd/mm/yy.

Table 8 Average ± standard deviation for ␦13 C (‰) of gas emitted from soil. All gas samples are bulk from rhizosphere and microbial sources. For Fields 11 & 14 in 2012, sources also included those from plants growing inside the chamber collar. p values are presented for t-tests. Field Pair

UTD

CTD

p value

2012 Field 1 & 2 Field 11 & 14

−21.08 ± 2.05 −26.78 ± 1.36

−21.59 ± 1.84 −23.86 ± 3.60

0.291 0.000

2013 Field 1 & 2 Field 11 & 14 Field 12 & 13

−19.01 ± 2.99 −25.02 ± 1.13 −24.78 ± 1.16

−19.01 ± 2.86 −24.57 ± 1.24 −25.21 ± 0.98

0.994 0.102 0.096

in soil moisture, the amount of CO2 produced and the gas diffusivity through the soil are affected, which may affect the magnitude of fractionation (Phillips et al., 2010). Soils with a high moisture content have reduced CO2 diffusion, reduced invasion of atmospheric CO2 , and greater rates of biological CO2 production thus resulting in greater CO2 concentrations and more depleted ␦13 C of emitted gas in relation to those associated with lower soil moisture (Cerling et al., 1991; Denny, 1993; Phillips et al., 2010). As demonstrated through greenhouse experiments and modeling, the ␦13 C of soil gas was more depleted in high soil moisture treatments compared to low soil moisture treatments, likely due to reduced gas transport in the soil media (Phillips et al., 2008, 2010). As such, it was expected that the reduced drainage in CTD fields would cre-

242

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243

ate higher soil moisture conditions causing reduced CO2 diffusion, reduced invasion of atmospheric CO2 , and greater rates of biological CO2 production thus resulting in greater CO2 concentrations and more depleted ␦13 C of the soil gas samples in relation to UTD fields where lower soil moisture was expected. While it was expected that CTD would alter residue decomposition and soil gas diffusivity through impacts on soil hydrology (e.g., decreased water table depth and increased soil water contents), similar SWC20 among CTD and UTD field pairs, as a result of hydrological factors previously described, apparently helped to also minimize differences in ␦13 C between 4 of the 5 CTD and UTD field pairs; at least in the surface soils. Notwithstanding that the average WTDBS for the growing seasons were significantly different among CTD and UTD field pairs (Fig. 4 and Table 2). For growing seasons with precipitation amounts within those studied herein, where differences in field hydrology (eg., surface soil water contents) resulting from CTD and UTD tile drainage management may have been greater (see Nangia et al., 2013; Sunohara et al., 2015, 2016), we would have expected more strongly expressed treatment differences with respect to ␦13 C in the sampled gas. Within each field, there is a trend of increasing contributions of the crops as a source for decomposition while the plants mature (Fig. 6). In field pair 1 and 2, 2013, the ␦13 C of the soil gas begins at its most depleted value (∼−25‰), which resembles the ␦13 C of the soil (SOM), and then becomes gradually more enriched, due to the increasing contribution of corn at ∼−12‰, until mid-July where it remains relatively stable (∼−17‰) for the remainder of the growing season. For soybean (Fields 1 & 2 in 2012, and Fields 11–14 in 2013), the ␦13 C of the soil gas is more enriched at the beginning of the season, reflecting the relatively enriched isotopic signatures of soil (∼−22 and ∼−25‰, respectively), and then becomes more depleted throughout the growing season as it received more inputs from depleted soybean (∼−27‰). At the end of the growing season, the soil gas signature becomes more enriched due to greater inputs from enriched soil. Enrichment trends could not be reliably determined for forage because of limited data. The ␦13 C of the gas samples was used to determine whether CTD management had a tile location effect on RT . It was expected, for reasons noted previously, that the location in a field with respect to a tile drain (directly above or between the tile drain laterals) would impact soil hydrology and subsequently residue decomposition and soil gas diffusivity. Similarly to bulk CTD vs UTD treatment comparisons previously described, mean values of gas ␦13 C did not differ between OT and BT locations for most CTD fields (p > 0.05). The exception was Field 11 in 2012, where BT gas samples had a more enriched ␦13 C signature compared to those for OT (Fig. 7). The observed difference in ␦13 C of OT and BT samples taken at locations within one CTD forage field could have been due in part to variations in emissions resulting from variable composition of alfalfa and timothy grass being present in the chambers. Alfalfa and timothy grass may have had different ␦13 C signatures and thus the proportion of alfalfa and timothy within chambers could therefore have effected the ␦13 C signature of the gas sampled.

4. Conclusion Agricultural BMPs may be used to address certain environmental issues, but environmental side effects/trade-offs and ancillary benefits, when possible, should be evaluated; as we don’t want to solve one problem while critically creating another. Controlled tile drainage (CTD) has historically been studied in the context of its potential to reduce nutrient loading to streams and boost crop yields. In this study, we explored how the practice might impact soil respiration and soil gas diffusivity. We expected that reduced tile drainage would create higher soil water contents resulting in

greater rates of biological CO2 production and more depleted ␦13 C of RT in relation to UTD fields. We further expected a more depleted ␦13 C of RT in CTD fields compared to UTD fields because of reduced CO2 diffusion and reduced exchange with atmospheric CO2 . However, ␦13 C of RT was not statistically different (p > 0.05) between four CTD and UTD field pair comparisons. It did differ for one CTD and UTD forage field pair, although a difference in biomass/plant type contribution in this case is suspected to be the cause. The generalized lack of difference in mean ␦13 C of RT between CTD and UTD field pairs may have resulted from their being insignificant differences in surface soil water contents among paired CTD and UTD fields due to the below and above normal precipitation conditions. During the drier 2012 season, water tables were often below the level at which tile drainage control could be employed, which likely engendered indifferences in surface soil hydrolgy among UTD and CTD field treatments. During the wetter 2013 field season, frequent rainfall over the growing season may have muted tile drainage treatment influences on surface soil water contents (i.e., surface soils for both treatment field systems had consistently high water contents; frequently wetted), resulting in no observable treatment effects. However, the isotopic approaches employed here could be tested on CTD/UTD systems for contrasting environmental/agronomic situations. It is noteworthy that this study was not conducted during the non-growing season, in part, because tile drainage control during that time in this region has potential to amplify environmental side-effects such as augmented phosphorus and sediment losses from fields (Ball-Coelho et al., 2012; Sunohara et al., 2016). However, in other regions of the world, such as in more temperate regions in the United States, CTD is practiced during these cooler and often wetter seasons where drainage control impacts on soil hydrology, soil respiration, and soil gas diffusivity may be more strongly expressed, in relation to summer growing season conditions.

Acknowledgements Financial support was provided by the South Nation Conservation as part of the Agricultural Greenhouse Gas Program (AGGP). The authors would like to acknowledge the landowners for their cooperation and support.

References Al-Kaisi, M.M., Yin, X., 2005. Tillage and crop residue effects on soil carbon and carbon dioxide emission in corn-soybean rotations. J. Environ. Qual. 34, 437–445. Askri, B., Ahmed, A.T., Abichou, T., Bouhlila, R., 2014. Effects of shallow water table, salinity and frequency of irrigation water on the date palm water use. J. Hydrol. 513, 81–90. Ball, B.C., Dobbie, K.E., Parker, J.P., Smith, K.A., 1997. The influence of gas transport and porosity on methane oxidation in soils. J. Geophys. Res. 102 (19), 23301–23308. Ball-Coelho, B.B., Murray, R., Lapen, D., Topp, E., Bruin, A., 2012. Phosphorus and sediment loading to surface waters from liquid swine manure application under different drainage and tillage practices. Agric. Water Manage. 104, 51–61. Brodie, C.R., Leng, M.J., Casford, J.S., Kendrick, C.P., Lloyd, J.M., Yongqiang, Z., Bird, M.I., 2011. Evidence for bias in C and N concentrations and ␦ 13C composition of terrestrial and aquatic organic materials due to pre-analysis acid preparation methods. Chem. Geol. 282 (3), 67–83. Cerling, T.E., Solomon, D.K., Quade, J., Bowman, J.R., 1991. On the isotopic composition of carbon in soil carbon dioxide. Geochim. Cosmoschim. Acta 55 (11), 3403–3405. Cheng, W., 1996. Measurement of rhizosphere respiration and organic matter decomposition using natural 13 C. Plant Soil 183, 263–268. Cicek, H., Sunohara, M., Wilkes, G., McNairn, H., Pick, F., Topp, E., Lapen, D.R., 2010. Using vegetation indices from satellite remote sensing to assess corn and soybean response to controlled tile drainage. Agric. Water Manage. 98 (2), 261–270. Clark, I.D., 2015. Groundwater Geochemistry and Isotopes. CRC press, Boca Raton, FL, 430 pp.

A. Van Zandvoort et al. / Agricultural Water Management 192 (2017) 232–243 De Silva, M.S., Nachabe, M.H., Simunek, J., Carnahan, R., 2008. Simulating root water uptake from a heterogeneous vegetative cover. J. Irrig. Drain. Eng. 134 (2), 167–174. Denny, M.W., 1993. Air and Water: the Biology and Physics of Life’s Media. Princeton university press, pp. 341. Drewitt, G., Wagner-Riddle, C., Warland, J., 2009. Isotopic CO2 measurements of soil respiration over conventional and no-till plots in fall and spring. Agric. Forest Meteorol. 149, 614–622. ECCC, 2015. Environment and Climate Change Canada, Russel Station, ON. Canadian Climate Normals 1981–2010, Date modified 2015-02-11; last date accessed 2015-10-14 http://climate.weather.gc.ca/climate normals/results 1981 2010 e.html?stnID=4376&autofwd=1. Elmi, A., Madramootoo, C., Hamel, C., 2000. Influence of water table and nitrogen management on residual soil NO3 − and denitrification rate under corn production in sandy loam soil in Quebec. Agric. Ecosyst. Environ. 79, 187–197. Fortin, M.C., Rochette, P., Pattey, E., 1996. Soil carbon dioxide fluxes from conventional and no-tillage small-grain cropping systems. Soil Sci. Soc. Am. J. 60 (5), 1541–1547. Gilliam, J.W., Skaggs, R.W., Weed, S.B., 1979. Drainage control to diminish nitrate loss from agricultural fields. J. Environ. Qual. 8 (1), 137–142. Gregorich, E.G., Ellert, B.H., Monreal, C.M., 1995. Turnover of soil organic matter and storage of corn residue carbon estimated from natural 13C abundance. Can. J. Soil Sci. 75, 161–167. Huang, L., 2015. Environment & Climate Change Canada, ASTD/Climate Research Division/CCMR, Toronto Canada. Original Data File: FRDG c13co2(09-10)QAQC PI (received on Oct. 2 2015). IPCC, 2017. Climate Change 2013. The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. In: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.). Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, p. 1535. Jin, Y., Jury, W.A., 1996. Characterizing the dependence of gas diffusion coefficient on soil properties. Soil Sci. Soc. Am. J. 60, 66–71. Kreba, S.A., Coyne, M.S., McCulley, R.L., Wendroth, O.O., 2013. Spatial and temporal patterns of carbon dioxide flux in crop and grass land-use systems. Vadose Zone J. 12 (4). Kross, A., Lapen, D.R., McNairn, H., Sunohara, M., Champagne, C., Wilkes, G., 2015. Satellite and in situ derived corn and soybean biomass and leaf area index: response to controlled tile drainage under varying weather conditions. Agric. Water Manage. 160, 118–131. Lal, R., 2004. Soil carbon sequestration to mitigate climate change. Geoderma 123 (1–2), 1–22. Li, H., Yi, J., Zhang, J., Zhao, Y., Si, B., Hill, R.L., Cui, L., Liu, X., 2015. Modeling of soil water and salt dynamics and its effects on root water uptake in Heihe arid wetland, Gansu, China. Water 7 (5), 2382–2401. Marshall, T.J., 1959. The diffusion of gases through porous media. J. Soil Sci. 10, 79–82. Mejia, M.N., Madramootoo, C.A., Broughton, R.S., 2000. Influence of water table management on corn and soybean yields. Agric. Water Manage. 46, 73–89. Moldrup, P., Olesen, T., Yoshikawa, S., Komatsu, T., Rolston, D.E., 2004. Three-porosity model for predicting the gas diffusion coefficient in undisturbed soil. Soil Sci. Soc. Am. J. 68, 750–759. Moldrup, P., Deepagoda, T.K.K.C., Komatsu, T., Kawamoto, K., Hamamoto, S., de Jonge, L.W., Rolston, D.E., 2013. Structure-dependent water-induced linear reduction model for predicting gas diffusivity and tortuosity in repacked and intact soil. Vadose zone J. 12 (3). Mordhorst, A., Perth, S., Horn, R., 2014. Influence of mechanical loading on static and dynamic CO2 efflux on differently textured and managed Luvisols. Geoderma 219–220, 1–13. Nangia, V., Sunohara, M.D., Topp, E., Gregorich, E.G., Drury, C.F., Gottschall, N., Lapen, D.R., 2013. Measuring and modeling the effects of drainage water management on soil greenhouse gas fluxes from corn and soybean fields. J. Environ. Manage. 129, 652–664. Ng, H.Y.F., Tan, C.S., Drury, C.F., Gaynor, J.D., 2002. Controlled drainage and subirrigation influences tile nitrate loss and corn yields in a sandy loam soil in Southwestern Ontario. Agric. Ecosyst. Environ. 90, 81–88. Nuutinen, V., Pöyhönen, S., Ketoja, E., Pitkänen, J., 2001. Abundance of the earthworm Lumbricus terrestris in relation to subsurface drainage pattern on a sandy clay field. Eur. J. Soil Biol. 37 (4), 301–304. O’Leary, M.H., 1988. Carbon isotopes in photosynthesis. Bioscience 38 (5), 328–336. Ontario Ministry of Food and Rural Affairs, 2009. Cereals: Fertility Management. OMAFRA, Guelph, ON, Canada (Accessed 4 Mar. 2015) http://www.omafra.gov. on.ca/english/crops/pub811/4fertility.htm.

243

Paustian, K., Andrén, O., Janzen, H.H., Lal, R., Smith, P., Tian, G., Tiessen, H., Van Noordwijk, M., Woomer, P.L., 1997. Agricultural soils as a sink to mitigate CO2 emissions. Soil Use Manage. 13, 230–244. Paustian, K., Six, J., Elliott, E.T., Hunt, H.W., 2000. Management options for reducing CO2 emissions from agricultural soils. Biogeochemistry 48, 147–163. Phillips, C.L., Nickerson, N., Risk, D., Kayler, Z.E., Ruch, W.D., Mix, A., Bond, B.J., 2008. Soil drying effects on the carbon isotope composition of soil respiration. In: Presented at American Geophysical Union Meeting, San Francisco, CA, December 15–19, 2008 (Accessed Feb 11 2016) http://cfpub.epa.gov/si/si public record Report. cfm?dirEntryId=199926&CFID=53081070&CFTOKEN=56573481. Phillips, C.L., Nickerson, N., Risk, D., Kayler, Z.E., Andersen, C., Mix, A., Bond, B.J., 2010. Soil moisture effects on the carbon isotope composition of soil respiration. Rapid Commun. Mass Spectrom. 24, 1271–1280. Rochette, P., Bertrand, N., 2008. Soil-surface gas emissions. In: Carter, M., Gregorich, E.G. (Eds.), Soil Sampling and Methods of Analysis. CRC Press, Boca Raton, FL, pp. 851–861. Rochette, P., Hutchinson, G.L., 2005. Measuring soil respiration in situ: chamber techniques. In: Hatfield, J.L., Baker, J.M. (Eds.), Micrometeorology in Agricultural Systems. ASA Monograph. No. 47. American Society of Agronomy, Madison, WI, pp. 247–286. Rochette, P., Flanagan, L.B., Gregorich, E.G., 1999a. Separating soil respiration into plant and soil components using analyses of the natural abundance of carbon-13. Soil Sci. Soc. Am. J. 63, 1207–1213. Rochette, P., Angers, D., Flanagan, L., 1999b. Maize residue decomposition measurement using soil surface carbon dioxide fluxes and natural abundance of carbon-13. Soil Sci. Soc. Am. J. 63, 1385–1396. Ryan, M.G., Law, B.E., 2005. Interpreting, measuring, and modeling soil respiration. Biogeochemistry 73, 3–27. Sánchez, M.L., Ozores, M.I., Colle, R., López, M.J., De Torre, B., García, M.A., Pérez, I., 2002. Soil CO2 fluxes in cereal land use of the Spanish plateau: influence of conventional and reduced tillage practices. Chemosphere 47 (8), 837–844. Schwen, A., Jeitler, E., Böttcher, J., 2015. Spatial and temporal variability of soil gas diffusivity, its scaling and relevance for soil respiration under different tillage. Geoderma 259–260, 323–336. Smedema, L.K., Rycroft, D.W., 1983. Land Drainage. Planning and Design of Agricultural Drainage Systems. B.T. Batsford Ltd., London. Smith, P., Martino, D., Cai, Z., Gwary, D., Janzen, H., Kumar, P., McCarl, B., Ogle, S., O’Mara, F., Rice, C., Scholes, B., Sirotenko, O., 2007. Agriculture. In Climate Change 2007. Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. In: Metz, B., Davidson, O.R., Bosch, P.R., Dave, R., Meyer, L.A. (Eds.). Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Sunohara, M.D., Craiovan, E., Topp, E., Gottschall, N., Drury, C.F., Lapen, D.R., 2014. Comprehensive nitrogen budgets for controlled tile drainage fields in eastern Ontario, Canada. J. Environ. Qual. 43, 617–630. Sunohara, M.D., Gottschall, N., Wilkes, G., Craiovan, E., Topp, E., Que, Z., Seidou, O., Frey, S.K., Lapen, D.R., 2015. Long-term observations of nitrogen and phosphorus export in paired-agricultural watersheds under controlled and conventional tile drainage. J. Environ. Qual. 44, 1589–1604. Sunohara, M.D., Gottschall, N., Craiovan, E., Wilkes, G., Topp, E., Frey, S.K., Lapen, D.R., 2016. Controlling tile drainage during the growing season in Eastern Canada to reduce nitrogen, phosphorus, and bacteria loading to surface water. Agric. Water Manage. 178, 159–170. Tuli, A., Hopmans, J.W., Rolston, D.E., Moldrup, P., 2005. Comparison of air and water permeability between disturbed and undisturbed soils. Soil Sci. Soc. Am. J. 69, 1361–1371. Wesstrom, I., Messing, I., 2007. Effects of controlled drainage on N and P losses and N dynamics in a loamy sand with spring crops. Agric. Water Manage. 87, 229–240. West, T.O., Marland, G., 2002. A synthesis of carbon sequestration, carbon emissions, and net carbon flux in agriculture: comparing tillage practices in the United States. Agric. Ecosyst. Environ. 91 (1), 217–232. Wicklund, R.E., Richards, N.R., 1962. Soil Survey of Russell and Prescott Counties. Rep No. 33 of the Ontario Soil Survey, Research Branch. Canada Dep. Of Agriculture, Ottawa.