Utilization of Wetlands for Agricultural Drainage Treatment: A Literature Review

Utilization of Wetlands for Agricultural Drainage Treatment: A Literature Review
By: Tamisan Latherow
October 2018

Grade B+
Abstract:

Wetlands are gaining popularity for use in agricultural waste treatments due to their natural ability to filter and fix excess nutrients, pesticides and herbicides, raising their ecosystem service valorization. With Nitrogen removal rates between 14-95% and Phosphorous removal rates between 25-95%, depending on design and infrastructure, wetlands provide a valuable tool for agricultural landscape management. However, those same services may be compromised if residence time and biodiversity of plants and microbial species are not properly managed. Mitigation efforts for restored wetlands surrounding agricultural landscapes must therefore be the primary goal of constructed wetlands for use in waste treatment and beneficial ecosystem services.

Introduction:

The world population is expected to reach 9.8 billion by 2050 (United Nations, 2017). Current estimates are that, barring all other considerations, 30% more food will need to be produced to feed everyone (Wezel, et al., 2014). To this end, many studies are focused on the creation of more food or the allocation of resources to manage supply chains, however, one must also consider the actual soil this food is to be grown on and the geospatial relationship to the surrounding landforms, catchments, and agricultural processes. How these systems interact and their influence down-stream directly impacts the efficiency of the lands’ ecosystem services, especially in wetlands. 

The Convention on Wetlands, aka. the Ramsar Convention, define wetlands as “areas of marsh, fen, peatland or water, whether natural or artificial, permanent or temporary, with water that is static or flowing, fresh, brackish or salt, including areas of marine water the depth of which at low tide does not exceed six metres” (Ramsar Convention Secretariat, 2016). According to Ramsar, wetlands cover roughly 6% of the Earth’s surface and ecosystem services of wetlands includes water and nutrient filtration and fixation, animal nurseries and habitats, storm surge protection, erosion and soil control, and natural locations of primary food stores such as rice and fish (Secretariat of the Convention on Biological Diversity, 2015). 

Wetlands have been incorporated into agricultural land use for generations and include traditional farming practices such as rice paddies, aquaculture, terraces, and reed production for fiber (Verhoeven & Settler, 2010). Much of the world’s wetlands have either been converted to and/or border agricultural lands (Davidson, 2014) and since they are typically at a lower elevation where the natural flow of catchment systems deposit soil from upstream they are often the recipients of fertilizer run-off (Yu, et al., 2018). 

Methods:

An analysis of wetlands and their relationship to agriculture was performed utilizing over twenty articles in published scientific journals across multiple disciplines (agriculture, ecology, economics, engineering, marine and freshwater ecosystems, botany and biology). These articles covered several different spatial scales, ranging from individual wetlands to national, regional, and global wetland overviews for the past few years. They also cover both natural (limited human alterations) and constructed (man-made) wetlands, hereafter referred to as NW or CW. The articles were broken down into three categories for further review: Nitrogen, Phosphorus, and herbicide and pesticide sequestration and then discussed under the subtopic of Ecosystem Services. All were reviewed at the landscape scale of application of agroecological practice and fall under the subheading of management of landscape elements (Wezel, et al., 2014).

For the purpose of this review, agroecological processes are those agricultural processes designed to produce significant amounts of food that also follow the natural cycles of the surrounding ecosystem for such items as nutrient cycling, natural pest management, soil and water conservation, biodiversity conservation, biological N fixation, carbon sequestration, and bacteria formation and management (Wezel, et al., 2014). To quantify the data around these topics, an agroecological framework is needed. 

The framework designed by Hill and MacRae (1996) was the base for this analysis with alterations by Wezel, et al. (2014). The framework focuses on efficiency increase, substitution practices, and redesign (Wezel, et al., 2014). Within these three pillars, Wezel, et al. identify specific practices related to crops and landscapes as well as three levels of management, those being the field level, farm level, and landscape level (Wezel, et al., 2014). We shall be looking at the agricultural landscape, specifically at if the addition of wetlands [either natural (NW) or constructed (CW)] for agricultural drainage treatment benefits or hinders the wetland’s natural ecosystem services. Wetlands were chosen due to their natural filtration ability and for the fact that 64-71% of wetlands have been lost since 1900 A.D. due to anthropogenic drivers such as agriculture (Davidson, 2014). 

Results:

Various studies (Dal Ferro, et al., 2018; Darwiche-Criado, et al., 2017; Ewing, et al., 2012; Kasak, et al., 2018) have been performed to identify the amount and efficiency of wetlands to manage fertilizer run-off from agricultural lands-Nitrogen (N) and Phosphorus (P) being the two predominant chemical elements; however various herbicides and pesticides have also been studied. With the growth of industrialized agriculture and the demand for more food, fertilizer treatments have increased. However, poorly -timed applications of these fertilizers account for much of the nutrient run-off into surrounding water sources and causes algae blooms, Eutrophication, and degradation of water quality (Kasak, et al., 2018). The wetland/catchment ratio creates two types of surface flow: in-stream and off-stream (Kasak, et al., 2018). The difference between in-stream (CWs located in the flow path of the drainage water, such as a river or ditch) and off-stream (where only part of the water flows through the CW) directly impacts the efficiency of the wetland to mitigate run-off levels. Generally, the ratio of wetland to catchment should be 0.5% with a relatively low flow velocity to allow the sedimentation of nutrients and soil run-off (Kasak, et al., 2018). 

Nitrogen and Phosphorus Sequestration

The three main ways nitrogen is taken up by an ecosystem is through vegetative uptake, sedimentation, and denitrification. Natural wetlands have been shown to retain around 64% of the Total Nitrogen loading (Saunders & Kalff, 2001), but in agricultural areas, TN appears mainly as nitrate, which is not as affected by sedimentation in high flow areas. In areas with lower flow or high plant coverage, where the roots of the plants allow for higher rates of denitrification and sedimentation and contribute to a higher residence time, an increase in the rate of retention can be found (Saunders & Kalff, 2001). Most agricultural associated wetlands are off-stream systems since water inundation leads to root rot in many crops while most aquaculture activities (such as rice farming) are conducted in-stream.  The natural formation of the wetland (eg. depressions, river banks, soil type, etc.) and the density of vegetation affects the sedimentation of solids and minimizes the transport of nutrients (Uwimana, et al., 2018). 

The increase in flow rate in in-stream systems minimizes the amount of time nutrients and soil has to settle and start filtering through the wetland soil stratus (Kasak, et al., 2018). In a NW this filtration happens through several layers of soil and aggregate before reaching the groundwater and is assisted by the various natural elements in the wetland such as cattails (Typhus), willows (Salix), and other plants, as well as microbial and fungi systems. In a CW, these same processes are managed by various liners and sedimentation ponds, some of which may be planted with vegetation. Most sedimentation ponds vary in the time water flows through the system from a few hours to a few weeks whereas a NW can take years to fully filter (Kasak, et al., 2018).

Nitrogen removal is dependent on many factors, but the most important are temperature, oxygen and carbon concentrations since they contribute to the various chemical (ammonification/volatilization) and microbial functions (nitrification/denitrification) (Kasak, et al., 2018). Factors such as groundwater seepage, plant-biomass N accumulation, and oxygen stratification within the water may affect the potential for N removal. These points were thought to contribute to the increase in NO₃-N concentration over the three-year study of the Kasak, et al. paper, where as the other papers saw a decrease in NO₃-N concentrations.

When evaluating nitrate-nitrogen loading capacity, the soil’s Total Carbon and pre-loading Total Nitrogen levels were also compared between organic and mineral soils. An earlier study by Ewing et al. (2012) found that organic soils had TC (29-35%) and TN (0.7–0.9%) while mineral soils had 6–7% and 0.2–0.3%, respectively (Messer, et al., 2017). This is important since the removal efficiency of N decreased when the TC/TN ratio approaches 5:1 (Messer, et al., 2017). The ratio is an indicator if enough oxygen occurred within the denitrification process, which accounts for 93% of nitrate uptake (Tournebize, et al., 2017). If there is a lack of oxygen, the conversion of nitrate into N2O (GHG) and N2 does not occur, causing overloads in the system. Flow rates and seasonality of rainfall and fertilizer application times, (Dal Ferro, et al., 2018) as well as the OM content of the wetland (Darwiche-Criado, et al., 2017), are direct contributors to this process. These issues can be mitigated by establishing a longer residence time within the CW system, allowing for more settling and the denitrification process to occur over a longer period. Diversified vegetation is also key since plants take up different amounts of nitrates over different parts of their life-cycle and a diversified vegetation profile within the wetland allows for the maximum amount (upwards of 7%) of nitrogen removal through this process  (Tournebize, et al., 2017). This is also assisted by the variation in the plant’s carbon storing capacity. For example, Nasturtium’s have been shown to support denitrification more fully than common reed (Phragmites) in wetland systems (Tournebize, et al., 2017). 

Likewise, Phosphorus retention is also dependent on water temperature and oxygen efficiency, microbial activity and plant uptake, however the retention rates for P average only around 14% when flow rates are at their lowest allowing for a longer standing residence time (flow rate of 5 L sˉ¹) (Kasak, et al., 2018). Of that 14%, 50-70% is found within the sediment and is considered as part of the permanent reservoir (Di Luca, et al., 2017). Plant uptake and microbial activity are directly linked to the pH, temperature and oxygen efficiency as plants and microbes are most active during warmer weather with access to appropriate oxygen stores for growth and development. Research by Johannesson, et al. (2017) shows a concern over other considerations such as peak flow times for grab sampling as differences in particulate phosphorous (PP) versus total phosphorus (TP) numbers and errors based on sampling during flow-events, that is, before, during, or after major flow events have arisen which could skew the amount of TP ratios (Johannesson, et al., 2017). These questions arise because of the seasonality in loading levels and their large impact on quantity of N and P in the system due to residence time and flushing of the wetlands.

In sediments, it has been shown that pH affects P absorption, as pH increases the absorption of phosphate decreases due to competition between absorption ratio of hydroxide (OH-) and soluble reactive phosphorus (the only form of P available for plant up-take) (Di Luca, et al., 2017). This decrease in oxidation-reduction potential (ability of water to cleanse or break-down waste) and the interference of OH- reduces the absorption of P into the sediment (Di Luca, et al., 2017). Since most P in wetlands is in insoluble organic or inorganic form (dominant ratios) (Bressler & Paul, 2015), the addition of nutrients into the CW and the subsequent loss of oxygen from Eutrophication can free phosphorus from the sediment and increase the TP available for plant uptake. However, the resulting change in C:P changes the microbial biomass and rate and variability in biodiversity within wetlands, which can lead to losses of biodiversity.

Herbicide and Pesticide Sequestration

Pesticides are often a factor after application and after heavy flow events such as storms; yet, they are often less than 0.5% of the applied application with rates rarely exceeding 3% (Tournebize, et al., 2017). This is three times less than found in nitrate levels, and is typically concentrated at the head of the in-flow region (up-stream of the watershed), however there is a wide variability in the transfer and transformation process surrounding each type of pesticide (Tournebize, et al., 2017).  The transfer process for pesticides from the water to the plant (28-55%) is of concern when the transfer moves not from water to plant, but from water to animals (especially marine animals) through tissue absorption (Tournebize, et al., 2017). There is also the issue of plant material where the chemicals have moved into the interior of the plant tissue through photoaccumulation and rereleased into the system upon decomposition (Tournebize, et al., 2017). Removal of these chemicals can sometimes be carried out via deabsorbion by flushing the system with more water, but it is not always effective. 

Transfer is not the only process pesticides can go through. New molecule creation from the dissolvement or degradation of the parent chemical can also occur through transformation and while the new chemicals are often in smaller quantities, they may be no less toxic. Transformation occurs predominantly in three ways within CWs: photodegradation (the effect of sun light), hydrolysis (via the movement of water), and biodegradation (through microbial processes) (Tournebize, et al., 2017). The primary agent of transformation is through microbial processes, specifically in aerobic conditions, where some chemicals such as the herbicide atrazine, widely used in corn and sugarcane production as a weed suppressant (Hayes, et al., 2010), have been shown to have mineralization rates upwards of 70-80% (Tournebize, et al., 2017). Mitigation strategies include permitting the water level to rise and drop to allow for oxygen to be starved from the system as well as sunlight to enter and assist in breaking down the chemicals further; this creates a combination of both reductive and oxidative conditions (Tournebize, et al., 2017). Less toxic chemical alternatives such as acetochlor and butachlor have been proven to degrade quicker in wetland environments (acetochlor in the rhizosphere and butachlor via microbial process) (Yu, et al., 2018), but the solubility of these chemicals is of concern. Acetocholr’s high water solubility makes it more prevalent in the water while butachlor’s low solubility leads to higher concentrations within the soil and subsoil. Simply changing to more easily degradable chemicals does not mitigate the effects if the ratios for chemical use increase; detection rates of 75-100% for acetochlor and 88.9-100% for butachlor within the surrounding wetlands were detected in the Chinese report (Yu, et al., 2018). Mitigation efforts to remove the herbicides are available and include increased growth of C. augustifolia plants to enrich soil microorganisms which can start to break down the chemicals. Butachlor can also be eliminated from the system by the removal of soil sediments (Yu, et al., 2018). 

In areas where organochlorines (OCPs) and DDT are used in agriculture, high concentrations of the OCPs have been found in tissue samples of fish species such as Tilapia and Catfish and have been linked to higher levels in secondary predators like birds and crocodiles, and even humans (Buah-Kwofie, et al., 2018). DDT use has been linked to eggshell thinning in aquatic birds and while the concentration levels are currently not at the level of human risk, according to EU regulations, further studies should be performed to find the threshold of consumed fish contaminants in the human population (Buah-Kwofie, et al., 2018).

Discussion:

Fig. 1. Ecological framework for landscape design on agricultural lands. “Arrows represent the main research axes and point towards the research frontiers. Darker shading indicates a greater level of current knowledge” (Landis, 2017).

While the methods for collecting data were similar across the research (water tests, biomass indexes, grab samples, etc.), percentages and ratios were not consistently presented making it difficult to evaluate across zones and papers. Future research would benefit from the creation and utilization of a general health assessment guide for wetlands such as the one created by the United States Environmental Protection Agency (Office of Water, U.S. Environmental Protection Agency, 2002). Generally, it was shown that wetlands do provide a valuable agricultural agency for the removal and dissipation of harmful agricultural runoff, as shown by the percentages in Table 1, however, there are key issues regarding if these same removals negatively affect the natural functionality of wetlands.

Table 1: Percentages of Removal and Plant Cover within Global Wetlands

Ecosystem Services Evaluation

The utilization of wetlands for waste treatment cannot be overstated, as it is clear from the research that their ability to remove large quantities of Nitrogen and Phosphorus as well as their potential for pesticide and herbicide removal is of great importance and benefit to the agricultural landscape. However, there are certain key components that should be noted and evaluated in further research including: temperature, seasonality, and pesticide management for denitrification purposes, buffer zones between wetlands and agricultural lands, plant biodiversity, and anthropogenic disturbances.
Denitrification effectiveness is a key aspect to the functionality of wetlands with temperature and dissolved oxygen being the two main components. High flow rates during flooding cause a slowing of denitrification and lack of oxygen into the system, whereas cycles of high water and low water with longer residence times and lower flow rates provide the greatest removal rates (Darwiche-Criado, et al., 2017). The combination of nitrates and pesticides within the system is a challenge for future studies, as certain chemicals (ex. Difenoconazole (fungicide),Deltamethrin (insecticide) and Ethofumesate (herbicide)) in high concentrations (500 mg/kg) have been shown to inhibit denitrifcation processes within the soil (Tournebize, et al., 2017).

Various studies (Buah-Kwofie, et al., 2018; Hayes, et al., 2010) have shown impacts of agricultural chemicals on the biodiversity of the down-stream systems including lakes and rivers. Fish and amphibian species provide a measure of the environmental exposure to the various chemicals due to their uptake structures: fish accumulate chemicals into their tissues and the amphibian’s skin allows for direct assimilation of chemicals via absorption. Fish have been shown to uptake 2-7% of pesticides into their flesh, up to 700% more than the surface soils these chemicals are actually sprayed on and chemicals like atrazine have been linked to the decline of global frog populations (Hayes, et al., 2010).

Mitigation strategies include planting crops that remove excess nitrates and phosphorus (such as corn) for several years between the primary agricultural zones and the wetlands (Ewing, et al., 2012) and planting narrow woodland buffer zones that would break up the flow of water and allow for greater species diversification of animals such as frogs (Sawatzky, et al., 2019). The two CW design methods (in-stream vs. off-stream) can also be utilized to mitigate each type of removal process. In-stream design methods that favor the removal of nitrates can be created where the CW is sufficiently broad allowing for generally equal spread of drainage water over the entire CW. Likewise, off-stream methods can be designed for pesticide removal. This is accomplished by the use of a gate to flood the land after pesticide use and direct the water into the CW. The gate is then closed after a set time allowing for higher residence time of effected waters in the CW (Tournebize, et al., 2017).

Plant biodiversity is another indicator of wetland health and ecosystem functionality. In a comparison between forested, agricultural, and urban wetlands, agricultural wetlands which were semi-permanently flooded with shallow water depth had the most biodiversity of plant species; 81.9% of which were herbs, ferns, or grasses (Moges, et al., 2017). Since plant uptake of N and P have been shown to account for 15-80% of Nitrogen and 24-80% of Phosphorus in the majority of research (Wu, et al., 2015), plant biodiversity and efficiency is of great concern. Anthropogenic factors such as harvesting, grazing, drainage, and waste treatment, while creating microhabitats, also disrupted the natural fauna and allowed for colonizer species which had the biggest effect on species diversification (Moges, et al., 2017). While this is of concern for seriously impaired wetlands, other ecosystem services have to be balanced for the good of the surrounding communities. In addition to aesthetic and recreational uses, many wetlands are used to support agricultural products and if taken into account with the activities, can greatly increase the perceived value of wetlands which may assist in their preservation or management. One such activity is mushroom production on common reed substrate harvested from wetlands.

It has been shown that Oyster Mushrooms grown on a reed substrate produce just as well if not better than on normal wood-shaving substrate and have the added benefit of removing small amounts of Nitrogen (2%), Phosphorus (0.2%), Potassium (1.34%), Pb and Cd from the wetlands through plant uptake (Hultberg, et al., 2018). The additional removal of heavy metals such as Lead (Pb) and Cadmium (Cd) assists with keeping the wetland water clearer for fish and more sensitive plants (Hultberg, et al., 2018). Another reed-focused activity is the creation of cattail (Typhus) pellets for use in wood burning stoves. These pellets have been shown to create between 7,266 and 8.551 BTU/lb which is comparable, if not better, than traditional wood pellets (7,266-7,739 BTU/lb) (Grosshans, et al., 2013) and convert roughly 0.7-0.11% Phosphorus, 0.79-1.53% Nitrogen, and 0.31-0.64% Potassium from the cattail’s dry weight (Grosshans, et al., 2015).

Overall, systems where traditional methods (that is non-fertilizer or chemical amendments) for farming that incorporate low-intensity crop production, grazing, and fish farming where the natural hydrology of the wetlands is left intact, may be the best choice for utilizing wetlands for both ecosystem services as well as economic viability (Verhoeven & Setter, 2010) for the surrounding communities.

Conclusion:

In conclusion, when looking at wetlands and their interaction with agricultural lands, ecology must co-exist with economics, and stakeholder engagement must be high for proper management of ecosystem services as well as economic viability. Wetlands used as landscape elements can benefit from low-intensity anthropogenic disturbances depending on temporal and spatial scales such as seasonality and mitigation efforts. The natural filtration ability of wetlands to remove fertilizers, pesticides and herbicides make constructing wetlands near agricultural lands highly economically viable, yet there are some key aspects of the ecosystem functions and biodiversity aspects that should be reviewed in more detail including the management of biodiversity within the wetlands for the greatest ecosystem service benefits. Overall, the research is clear that wetlands can be used for agricultural drainage treatment, but there is more uncertainty around if treatments negatively impact the very ecosystem services that those agricultural lands are counting on. Incorrect chemical loads and seasonal variability can greatly impact the efficiency of wetlands and should be taken into account when designing or utilizing CWs. The agroecological functionality of wetlands are tied to its hydrology, geomorphology, and biodiversity and these need to be key aspects of evaluation before utilization. 

Bibliography:

Bressler, D. W. & Paul, M. J., 2015. Effects of Eutrophication on Wetland Ecosystems, s.l.: Tetra Tech, Inc..

Buah-Kwofie, A., Humphries, M. S. & Pillay, L., 2018. Bioaccumulation and risk assessment of organochlorine pesticides in fish from a global biodiversity hotspot: iSimangaliso Wetland Park, South Africa. Science of the Total Environment, Volume 621, pp. 273-281.

Dal Ferro, N., Ibrahim, H. M. S. & Borin, M., 2018. Newly-established free water-surface constructed wetland to treat agricultural waters in the low-laying Venetian plain: Performance on nitrogen and phosphorus removal. Science of the Total Environment, 639(1), pp. 852-859.

Darwiche-Criado, N. et al., 2017. Effects of wetland restoration on nitrate removal in an irrigated agricultural are: The role of in-stream and off-stream wetlands. Ecological Engineering, Volume 103, pp. 426-435.

Davidson, N., 2014. How much wetland has the world lost? Long-term and recent trends in global wetland area. Marine and Freshwater Research.

Di Luca, G. et al., 2017. Phosphorus distribution pattern in sediments of natural and constructed wetlands. Ecological Engineering, Volume 108, pp. 227-233.

Ewing, J., Vepraskas, M., Broome, S. & White, J., 2012. Changes in wetland soil morphological and chemical properties after 15, 20, and 30 years of agricultural production. Geoderma, Volume 179-180, pp. 73-80.

Grosshans, R. et al., 2013. Cattail Harvesting for Carbon Offsets and Nutrient Capture: A "Lake Friendly" Greenhouse Gas Project, Winnipeg: International Institute for Sustainable Development.

Grosshans, R. et al., 2015. Cattail Biomas in Watershed-Based Bioeconomy: Commercial-scale harvesting and processing for nutrient capture, biocarbon and high-value bioproducts, Winnipeg: International Institute for Sustainable Development.

Hayes, T. B. et al., 2010. Atrazine induces complete feminization and chemical castration in male African clawed frogs (Xenopus laevic). Proceedings of the National Academy of Sciences of the United States, 107(10), pp. 4612-4617.

Hill, S. & MacRae, R., 1996. Conceptual Framework for the Transition from Conventional to Sustainable Agriculture. Journal of Sustainable Agriculture, 7(1), pp. 81-87.

Hultberg, M. et al., 2018. Adding benefit to wetlands-Valorization of harvested common reed through mushroom production. Science of the Total Environment, Volume 637-638, pp. 1395-1399.

Johannesson, K. M., Tonderski, K. S., Ehde, P. M. & Weisner, S. E., 2017. Temporal phosphorus dynamics affecting retention estimates in agricultural constructed wetlands. Ecological Engineering, Volume 103, pp. 436-445.

Kasak, K., Kill, K., Parn, J. & Mander, U., 2018. Efficiency of newly established in-stream constructed wetland treating diffuse agricultural pollution. Ecological Engineering, 119(1), pp. 1-7.

Landis, D. A., 2017. Designing agricultural landscapes for biodiversity-based ecosystem services. Basic and Applied Ecology, Volume 18, pp. 1-12.

Messer, T. L. et al., 2017. Nitrate removl potential of restored wetlands loaded with agricultural draignage water: A mesocosm scale experimental approach. Ecological Engineering, Volume 106, pp. 541-554.

Moges, A. et al., 2017. Plant species composition and diversity in wetlands under forest, agriculture and urban land uses. Aquatic Botany, Volume 138, pp. 9-15.

Office of Water, U.S. Environmental Protection Agency, 2002. Methods for Evaluating Wetland Condition: Introduction to Wetland Biological Assessment, Washington, D.C.: The U.S. EPA.

Ramsar Convention Secretariat, 2016. RAMSAR, Gland: s.n.

Saunders, D. & Kalff, J., 2001. Nitrogen retention in wetlands, lakes and rivers. Hydrobiologia, Volume 443, pp. 205-212.

Sawatzky, M. E., Martin, A. E. & Fahrig, L., 2019. Landscape context is more important than wetland buffers for farmland amphibians. Agriculture, Ecosystems and Environment, Volume 269, pp. 97-106.

Secretariat of the Convention on Biological Diversity, 2015. Wetlands and Ecosystem Services, Montreal: s.n.

Tournebize, J., Chaumont, C. & Mander, U., 2017. Implications for constructed wetlands to mitigate nitrate and pesticide pollution in agricultural drained watersheds. Ecological Engineering, Volume 103, pp. 415-425.

United Nations, 2017. World Population Prospects – 2017 Revision: Global population. [Online] 

Available at: https://www.un.org/development/desa/publications/graphic/wpp2017-global-population

[Accessed 9 October 2018].

Uwimana, A., van Dam, A. A., Gettel, G. M. & Irvine, K., 2018. Effects of agricultural land use on sediment and nutrient retention in valley-bottom wetlands of Migina catchment, southern Rwanda. Journal of Environmental Management, Volume 219, pp. 103-114.

Verhoeven, J. & Settler, T., 2010. Agricultural use of wetlands: opportunities and limitations. Annals of Botany, 105(1), pp. 155-163.

Verhoeven, J. T. & Setter, T. L., 2010. Agricultural use of wetlands: opportunities and limitations. Annals of Botany, Volume 105, pp. 155-163.

Wezel, A. et al., 2014. Agroecological practices for sustainable agriculture. A review. Agronomy for Sustainable Development, 34(1), pp. 1-20.

Wu, H. et al., 2015. A review on the sustainability of constructed wetlands for wastewater treatment: Design and operation. Bioresource Technology, Volume 175, pp. 594-601.

Yu, X. et al., 2018. Herbicide accumulations in the Xingkai lake area and the use of restored wetland for agricultural drainage treatment. Ecological Engineering, Volume 120, pp. 260-265.