Enhancing environmental sustainability in eastern Canada's corn agroecosystem with controlled drainage and subsurface irrigation

The utilization of irrigation and drainage systems in agriculture constitutes a crucial component in increasing water use efficiency and agricultural productivity. However, agricultural production-related environmental issues originating from irrigation and drainage may cause adverse environmental impacts. These detrimental environmental impacts are predominantly recognized as non-point pollution, including soil erosion, diminishing carbon sequestration of farmland, deterioration of water quality, and greenhouse gas (GHG) emissions (King et al. 2015, 2018; Plach et al. 2018). Research scholars also highlight the importance of understanding irrigation water quality and its impact on long-term productivity, as well as the impact of climate change (reduced rainfall irrigation in particular) on crop productivity (Çadraku 2021; Jabal et al. 2022).

The eastern Canadian region has witnessed the widespread adoption of tile drainage for draining the farmland during the spring and fall, due to the low terrain and relatively high precipitation (Morrison et al. 2014)¹. However, free drainage (FD) may operate as pipes for the transportation of nitrates into open ditches and surface waters, increasing the concentration of chemicals (mainly nutrients and pesticides) in agricultural drainage water (Tanji & Keyes 2002). The other environmental risks are also aggravated by this drainage practice, including ammonia volatilization, nitrification and denitrification, and GHG emissions (Kanter et al. 2021). Controlled tile drainage and subsurface irrigation (CDSI) has been developed to address the setbacks of conventional drainage systems and improve agricultural water management. CDSI represents an innovative water management solution that can be retrofitted to FD systems, facilitating a flexible structure at the tile outlet². The outlet can be elevated to reserve water in the soil profile during the summer and lowered before planting (and again before harvest) to manage outflow and maintain the desired water table, thereby ensuring a recycled water supply for crop growth.

The implementation of CDSI has shown significant agronomic and environmental benefits (Madramootoo et al. 1997; Rasouli et al. 2014). Controlled drainage is associated with 4% yield increases for corn and 3% yield increases for soybean crops in the St. Lawrence lowlands region of eastern Canada (Crabbé et al. 2012). In addition, studies have shown that CDSI can significantly reduce the annual loads and flow-weighted concentrations of nitrogen (Ritter et al. 1995; Elmi et al. 2004), retain soil nutrients (Ballantine & Tanner 2013), and decrease in-stream water phosphorus (P) loads (Bishop et al. 2005). Although CDSI has been found to decrease total GHG fluxes slightly, CH₄ fluxes may increase strongly³ (Van Huissteden et al. 2006). Recent research has focused on the effects of agricultural water management on GHG emissions (Dinsmore et al. 2009; Hooijer et al. 2010; Musarika et al. 2017; Hagedorn et al. 2022).

The topography and soil type of farmland in eastern Canada offer favourable conditions for the adoption of CDSI, particularly in areas with poor drainage⁴ (Land Improvement Contractors of Ontario 2008). Nevertheless, the implementation of CDSI faces obstacles due to limited research on its comprehensive environmental benefits. The absence of comprehensive environmental assessments curtails the knowledge of potential benefits and trade-offs of implementing CDSI, discouraging the investment and supportive policies for CDSI. In response to that, this study employed life cycle assessment (LCA) to assess the entire life cycle environmental impacts of a corn agroecosystem under CDSI compared to the status quo of FD in eastern Canada. By comparing the environmental performance of CDSI and FD in the corn agroecosystem, this study aims to provide insight into the potential environmental benefits and/or trade-offs of implementing CDSI. Consequently, this comprehensive assessment may support producers' investment of CDSI and further adoption of beneficial water management practices, concerning the interest of policymakers and other stakeholders.

The rest paper is structed as follows. Section 2 provides a detailed description of the study's methodology, including the research site and the steps taken to conduct a comparative LCA of a corn production system under CDSI and FD. Section 3 presents the results and discussions of the study, with a particular focus on comparing the environmental performance of CDSI and FD. Finally, in Section 4, the conclusions of the study are presented.

This study builds upon long-term field experiments on a 4.2 ha field located in St-Emmanuel (Côteau-du-Lac), Quebec, to investigate the impacts of innovative agricultural water management systems. The topography of the research site is flat, with a slope of around 0.5% (Tait et al. 1995). Recent increases in drought conditions, characterized by low rainfall and high evaporation during the summer months, have been observed in the region. Historical monthly precipitation data from Environment and Climate Change Canada (2022) indicate that, from 1993 to 2017, rainfall from May to September was below the 40-year average for most years, with three consecutive years of drought occurring from 1993 to 1995, 2001 to 2004, and 2007 to 2010. These reduced precipitation levels highlight the need for irrigation. However, inadequate irrigation installations in this region have exposed the risks and vulnerabilities of agricultural production to climate change. For example, according to Census of Agriculture (2021), less than 10% of farmland was irrigated in Côteau-du-Lac in 2021, and only 2% in Quebec. The current state of irrigation infrastructure implies the potential benefits of implementing agricultural water management in this region to sustain agriculture under climate change.

In 2014 and 2015, the comparative experiments were conducted for two water management systems in split plots. The base scenario was FD, placing subsurface drainpipes of 76 mm diameter at the centre of each plot and 1 m below the soil surface. The treatment scenario was CDSI, which involved the installation of an excessive manual control structure to manage the water outflow. This device enables the manual addition or removal of stoplogs to maintain the desired water table level. Further details regarding the field experiments and raw data collection can be found in Crézé & Madramootoo (2019).

This study compares environmental impacts of corn production with conventional drainage technology and CDSI using LCA. Specifically, the LCA focused on the cropping system under a new agricultural water management device, which upgraded the base technology to address agricultural sustainability. In other words, this study investigated the differences in environmental impacts between CDSI and FD technologies during the corn production process, which has not been explored in previous agricultural LCA that focus on the environmental outcome of the whole production process.

Data collection is crucial for developing a life cycle inventory that can assess regional differences in the environmental impacts of corn grain production (Boone et al. 2016). On-field nitrogen and pesticide emissions from the cultivation of the crop production systems should be precisely measured. However, because LCA was primarily established for industrial products and operations before being expanded to agriculture, it confronts challenges in investigating the environmental impacts of agricultural production processes (Caffrey & Veal 2013). Agricultural production processes often involve multiple products from a single cropping system, farm management practices, and new agricultural technologies, which can make it difficult to thoroughly examine them in agricultural LCA. Therefore, the foundation of developing agricultural LCA lies in a sufficient and reasonable database that includes all the essential processes of agricultural production, especially agricultural emissions data.

In this study, LCI data were sourced from both field experiments and measurements and existing agricultural LCI databases. This study used raw data and coefficient estimation of inputs to quantify the environmental inventory flows, and to measure and calculate emissions from crop production systems to soil, water, and air.

In terms of field data, this study collected operational and biophysical data during the 2014 and 2015 growing seasons, including fertilizer, herbicide, water use, grain yields and GHG emissions as primary data from corresponding experimental plots with CDSI and FD. Furthermore, emissions to soil and water were collected from a series of earlier studies comparing the environmental impacts of CDSI to FD at the same research site in earlier years, such as soil nitrate-N (NO₃-N) pollution and drainage water NO₃-N concentrations (Elmi et al. 2004), and phosphorus pollution in drainage water (Hebraud 2006; Valero et al. 2007). The raw emission data for corn production under the CDSI and FD were directly measured from farming experiments and sampling. For example, GHG samples were collected from each experimental spot using field-deployed, vented non-steady-state gas chambers to quantify weekly soil N₂O, CO₂, and CH₄ fluxes (Crézé & Madramootoo 2019). Additionally, a random-effects model was used to control for the influence of different temperatures and soil moisture on GHG emissions between 2014 and 2015. Daily GHG fluxes were weighed using a 40-year average temperature and precipitation. Thus, the inconsistency of climatic conditions between 2014 and 2015 did not reflect in the assessed GHG emissions of corn production under CDSI versus FD controlling time effects (2 years) and individual effects (two technologies). Although direct on-field sampling can only be applied to a few spot measurements due to high research expenditure, it provides a validated primary source for the comparable GHG emissions of corn production under two water management technologies.

This study also utilized the World Food LCA Database (WFLDB) v3.0 (Nemecek et al. 2014) and AGRIBALYSE v1.3 (Koch & Salou 2016) to supplement materials and input processes that could not be gathered from field measurements. LCI databases developed in various research formed the basis of a biophysical inventory of corn grain production based on comparable flows. Data from the agricultural LCI databases used in this study contained: (1) NH₃ emissions, (2) heavy metals emissions to agricultural soil, surface water and groundwater, (3) pesticide emissions, and (4) machinery for field operations. Notably, the origin of the supplemental databases was European, thereby limiting their application to different regions. Nevertheless, the use of secondary LCI databases and modeling techniques did not compromise the credibility and accuracy of the study's comparable LCA results, because of this study's reliance on validated data for assessing environmental impacts of corn production under CDSI and FD. Subsequently, input and output flows were established and measured in functional units following the quantitative analysis of resources, energy, and environmental emissions within the pre-defined life cycle boundaries.

The life cycle impact assessment (LCIA) method used in this study was CML 2001, an intermediate point environmental impact analysis method issued by the Environmental Research Center of Leiden University (Guinée 2001). The study calculated the environmental impacts of corn production with CDSI and FD based on CML 2001 impact categories. To assess the net environmental impacts of CDSI implementation, the study compared the environmental impacts of corn production under these two water management systems. These environmental impact categories included land use (measured as Soil Organic Carbon (SOC)), climate change by CO₂ equivalent, eutrophication potential by PO₄³⁻, acidification potential by SO₂ equivalent, human toxicity and ecotoxicity by 1,4 dichlorobenzenes (1,4-DCB).

Environmental impacts were compared for corn production from the 2014–2015 growing seasons with CDSI and FD, including climate change, eutrophication potential, acidification potential and toxicity. The results are reported in Table 1, using the 1-tonne corn grain as the functional unit. Particularly, yearly variations of assessed life cycle impacts of climate change, eutrophication, and acidification are shown in Table 2.

In terms of land use, corn production with CDSI resulted in an average of 2.55% compared to FD during 2014 and 2015 growing seasons. This reduction can be attributed to the increased yields resulting from the implementation of CDSI, which required the same amount of land resources as FD. This finding underscores the importance of crop yields in assessing agricultural water management systems.

Loss of soil productivity, such as erosion and degradation, can lead to lower crop yields and higher production costs (Pimentel et al. 1995). Soil water retention and organic carbon are essential to agronomic productivity (Lal 2020). Thus, CDSI implementation can help conserve soil organic carbon by adjusting soil moisture content instead of allowing soil runoff by FD. Additionally, net removals from cropland play a significant role in GHG emissions mitigation. In Canada, current net removals from cropland in 2019 are 4.2 Mt, 6.2 Mt lower than in 2005 (Environment & Climate Change Canada 2019). The declining net removals of GHG emissions from cropland emphasize the need to reverse the trend of soil deterioration. In this context, the implementation of CDSI has been found to utilize fewer land resources while enhancing carbon sequestration in farmland.

The assessed climate change impacts of producing 1-tonne corn with CDSI and FD in 2014 were 222 and 311 kg, respectively. In 2015, climate change impacts were 283 and 204 kg CO₂-eq. for the corn production system with CDSI and FD. The total GHG emissions from corn production system in this study were about 3,000 kg CO₂-eq. ha⁻¹. This amount corresponds to a low-medium GHG emissions threshold in Farrell et al. (2006), which discovered that GHG emissions associated with corn (Zea mays L.) production ranged from 2,441 to 4,201 kg CO₂-eq. ha⁻¹ year⁻¹ by analyzing results from six literature works.

CDSI indicated a modest reduction in climate change impacts for the two growing seasons. The cumulative variation of climate change impacts between corn production with FD and CDSI was 6.25 kg CO₂-eq. t⁻¹ year⁻¹. In 2014, the life cycle climate change impacts with FD were 40% lower than that with CDSI. However, in 2015, the reverse was observed, as CDSI decreased climate change impacts by 28% relative to FD. Analysis of the variation in the same water management technology between the two growing seasons shows that GHG emissions from the corn agroecosystem with CDSI dropped sharply in 2015, by 107 kg CO₂-eq. t⁻¹ year⁻¹, accounting for 34% of the GHG emissions from the 2014 CDSI production system. In contrast, total GHG emissions for FD increased by 61 kg CO₂-eq. t^–1 of corn production in 2015, a 27% rise from 2014. Overall, the corn agroecosystem with CDSI in the growing season of 2015 reached the lowest climate change impacts. The implementation of CDSI resulted in a minor reduction of 2.43% compared to FD in the continuous corn production system during the 2014 and 2015 growing seasons.

Life cycle eutrophication potential impacts of corn production reduced from 0.386 kg PO₄-eq. t⁻¹ with FD to 0.235 kg PO₄-eq. t⁻¹ with CDSI in 2014 and 0.355 kg PO₄-eq. t⁻¹ of FD to 0.151 kg PO₄-eq. t⁻¹ of CDSI in 2015, which decreased by 39 and 58% respectively. The environmental performance of CDSI is consistent across the two growing seasons suggest its implementation is an effective measure for reducing the eutrophication potential impacts of corn production.

The corn production with CDSI vastly increased acidification impacts in 2014. However, in 2015, implementing CDSI caused a 32.15% reduction in the acidification potential impacts compared to FD. Overall, the corn production system with CDSI showed 20% more significant environmental impacts than the conventional FD in the acidification category. The results indicate that the implementation of CDSI may pose a risk in increasing acidification potential impacts. However, the favorable environmental performance of CDSI in the 2015 growing season drives the further investigation into the cause of the inconsistency. Also, the development of proper user guidelines is necessary to improve the environmental performance of CDSI in the acidification category.

The environmental impact measurements of corn production with CDSI were 0.176, 0.079, and 2.845 kg 1,4-DCB-eq. t⁻¹ for human toxicity, terrestrial ecotoxicity, freshwater aquatic ecotoxicity, respectively. On average, there was a 2.55% reduction in toxicity compared to FD, attributable to the mitigation of herbicide and pesticide per unit, resulting from increased yields due to CDSI implementation.

The results are compared to other agricultural LCA studies of the corn production system to explain the study's reliability. Despite using different LCA methods and research regions, these comparative studies provide an environmental impact range associated with this research. Given the similarity in research region (both in North America) and crop production system (corn grain production), this study employed the results from Kim et al. (2009) to compare climate change, eutrophication and acidification impacts on corn grain production.

In Kim et al. (2009), GHG field emissions associated with corn grain production ranged from 127 to 625 kg CO₂-eq. t⁻¹ year⁻¹, which is consistent with the climate change impacts assessed in this study. Moreover, according to Kim et al. (2009), eutrophication and acidification associated with corn grain production in US ranged from 0.5 to 2.0 kg PO₄-eq. t⁻¹ and 1.5 to 5.8 kg SO₂-eq. t⁻¹, respectively, which are 5–10 times higher than those reported in this study. The discrepancy is partly due to differences in inventory methods used to estimate the emissions. This study used field experiment samples, whereas Kim et al. (2009) employed a simulation model to estimate N₂O and N₂ gas emissions from soils resulting from denitrification. Additionally, the study involved water management in the LCA, which was not done in the previous research. Even the base scenario, conventional tile drainage, shows benefits in improving water and soil quality compared to the no-tile drained, demonstrating the environmental benefits of agricultural water management in sustaining agricultural production.

This study emphasizes the environmental benefits of CDSI implementation on water quality and climate change based on the technology's design functionality. As North America experiences a trend of reduced precipitation and increased irrigation demand, adapting conventional irrigation and drainage to handle the upcoming challenge of climate change is crucial (Lobell et al. 2008). This is particularly important in the context of growing societal and environmental obligations to develop sustainable agroecosystems with limited natural resources.

Moreover, this study enlightens the potential interrelationships among different environmental impact categories, including co-benefits and trade-offs revealed from biophysical indicators across the assessed categories. For instance, co-benefits reflect on that GHG mitigation can also reduce acidification potential due to fewer N₂O emissions. Nevertheless, trade-offs exist between climate change and eutrophication because reduced N-leaching that may convert into potential N₂O emissions. In other words, the trade-offs imply that CDSI still holds the possibility of transforming decreased N-leaching into added N₂O emissions, thereby generating unfavorable environmental influences on soil and water. The interactions between diverse environmental impacts underscore the need to consider the agroecosystem when assessing CDSI. Also, agricultural water management can incorporate other environmentally beneficial practices (e.g., fertilizer application, no-till or reduced-till farming) to achieve agri-environmental improvements. Thus, as an innovative technology and its associated beneficial management practices, CDSI should facilitate agricultural water management to mitigate the comprehensive environmental impacts in crop production.

This study compares the environmental impacts of the corn production system under controlled drainage with sub-irrigation (CDSI) and conventional tile drainage (FD) in a corn agroecosystem during the 2014–2015 growing seasons in Quebec, eastern Canada. Through LCA, this study evaluates the environmental performance of CDSI as an innovative water management system.

The life cycle impact results show that CDSI had a positive effect on reducing environmental impacts of corn production compared to FD, in the aspects of land use, climate change, eutrophication potential, human toxicity, terrestrial ecotoxicity, and freshwater aquatic ecotoxicity. However, CDSI show underperformance in potential acidification impacts due to higher N₂O emissions, also affecting the climate change impacts in 2014. CDSI generally shows positive impacts on reducing GHG emissions, with consistent CO₂ emissions regarding the climate change impacts category. Moreover, the eutrophication potential impacts resulting from implementing CDSI were significantly reduced due to vastly lowering N and P emissions to water compared to FD. The interactions between various environmental impacts sheds light on the importance of combining water management and fertilizer application. However, improper fertilizer application can cause acidification and eutrophication impacts, along with impacts on N₂O emissions in climate change. Hence, reducing N₂O emissions from CDSI is crucial for comprehensive environmental benefits, and further research is needed to interpret the inconsistency of N₂O emissions for corn agroecosystems with CDSI.

Policy ensuring environmental responsibility in the context of climate change for the society can benefit stakeholders and promote sustainable economic development (Hang 2022). This study has important policy implications, as it provides a comprehensive assessment of CDSI implementation for mitigating agricultural environmental impacts through water management. The results contribute to scientific understanding of environmental impacts of transiting to corn agroecosystem with CDSI assessed on a biophysical basis. Moreover, this study can support further monetary measurement of socio-economic evaluation on CDSI, such as cost–benefit and multi-criteria analysis. Currently, inadequate interest to producers in adopting CDSI due to low investment return of farmland water management infrastructure (Marmanilo et al. 2021). However, the environmental assessment in this study highlights the potential benefits of CDSI. Stakeholders, including producers, researchers, technology manufacturers, and policymakers, can refer to environmental benefits and the following monetary evaluation of CDSI to encourage the adoption of CDSI-based water management practices. Consequently, the related agri-environmental policy supporting CDSI implementation may help to fulfill the purpose of reducing the overall agri-environmental impacts.

This research was supported by Agriculture and Agri-Food Canada's Agricultural Greenhouse Gases Program (AGGP) [grant number AGGP2-020]. The authors wish to thank Mfon Essien of McGill University for structuring farm data and reviewing a previous draft of this manuscript.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.