Advances in the simulation of nutrient dynamics in cold climate agricultural basins: Developing new nitrogen and phosphorus modules for the Cold Regions Hydrological Modelling Platform

https://doi.org/10.1016/j.jhydrol.2021.126901Get rights and content

Highlights

  • Application of popular catchment nutrient models is problematic in cold regions.

  • New nutrient modules have been developed for the Cold Regions Hydrological Model.

  • The model was applied to a sub-basin of the increasingly eutrophic Lake Winnipeg, Canada.

  • Simulated SWE, discharge, NO3, NH4, SRP and partP were compared against observations.

  • Typical ∼9 day-freshet accounted for 16–31% of the total annual nutrient load.

Abstract

Excess nutrients in aquatic ecosystems is a major water quality problem globally. Worsening eutrophication issues are notable in cold temperate areas, with pervasive problems in many agriculturally dominated catchments. Predicting nutrient export to rivers and lakes is particularly difficult in cold agricultural environments because of challenges in modelling snow, soil, frozen ground, climate, and anthropogenic controls. Previous research has shown that the use of many popular small basin nutrient models can be problematic in cold regions due to poor representation of cold region hydrology. In this study, the Cold Regions Hydrological Modelling Platform (CRHM), a modular modelling system, which has been widely deployed across Canada and cold regions worldwide, was used to address this problem. CRHM was extended to simulate biogeochemical and transport processes for nitrogen and phosphorus through a complex of new process-based modules that represent physicochemical processes in snow, soil and freshwater. Agricultural practices such as tillage and fertilizer application, which strongly impact the availability and release of soil nutrients, can be explicitly represented in the model. A test case in an agricultural basin draining towards Lake Winnipeg shows that the model can capture the extreme hydrology and nutrient load variability of small agricultural basins at hourly time steps. It was demonstrated that fine temporal resolutions are an essential modelling requisite to capture strong concentration changes in agricultural tributaries in cold agricultural environments. Within these ephemeral and intermittent streams, on average, 30%, 31%, 20%, and 16% of the total annual load of nitrate (NO3), ammonium (NH4), soluble reactive phosphorus (SRP), and particulate phosphorous (partP)NO3, NH4, SRP and partP occurred during the episodic snowmelt freshet (9 days, accounting for 21% of the annual flow), but shows extreme temporal variation. The new nutrient modules are critical tools for predicting nutrient export from small agricultural drainage basins in cold climates via better representation of key hydrological processes, and a temporal resolution more suited to capture dynamics of ephemeral and intermittent streams.

Introduction

Reducing nutrient losses from agricultural fields has been a major priority worldwide for many years due to increasing concerns with enhanced aquatic productivity and algal blooms. Water quality models for both basin and in-stream studies have been widely used to support nutrient management, but have often been problematic in seasonally cold regions such as Canada and the northern United States due to deficiencies in the representation of key processes specific to these regions. Cold regions hydrology cannot be represented by the classical concepts of rainfall-runoff models due to water storage by the seasonal snowcover, snow redistribution by wind, radiation-driven snowmelt, infiltration to and runoff over seasonally frozen ground, poorly defined drainage due to glacial geomorphology, and highly episodic runoff events (Pomeroy et al., 2007). Regional biogeochemistry in soils and runoff is challenging to model due to cold temperatures and seasonal soil freezing that influence nutrient release from soil–plant systems, plant uptake and microbial activity, which in combination with management practices (including fertilizer applications, tillage practices and wetland drainage) affect the hydrochemistry of soils and runoff (Baulch et al., 2019, Costa et al., 2020a, Irvine et al., 2019, Van Esbroeck et al., 2017, Macrae et al., 2007).

The dynamics of nutrient storage and release in cold climates are strongly affected by various cold regions hydrological processes and conditions (Deelstra et al., 2009). Snowpacks collect and transform chemicals during winter and rapidly release them during snowmelt (Pomeroy et al., 2005), with a significant portion of the nutrients contained in runoff being transformed and retained in topographical depressions (Neely and Baker, 1989, Crumpton and Isenhart, 1993, Birgand et al., 2007). Spring snowmelt is the largest runoff event of the year in cold regions such as the Northern Great Plains of North America (Gray et al., 1970), and accounts for most of the annual nutrient export (Baulch et al., 2019). The magnitude of peak flows furing spring freshet depends not only on overwinter snow accumulation but also on the antecedent soil moisture and basal snowpack and ground ice conditions (Gray et al., 1986, Pomeroy et al., 2007). Except for runoff from intensive convective rainfall events, summer flows are often small (Gray et al., 1970, Pomeroy et al., 2007).

Nitrogen (N) and phosphorus (P) transported via cold regions agricultural runoff originate in soil, vegetation, or to a lesser extent, the snowpack. The soil N pool is highly dynamic with weathering of soil parent material and decomposition of soil organic matter providing sources of mineral N (NO3-N and NH4-N) at rates depending on soil type and climate. Additional N enters the landscape through fertilizer application, plant residues, and atmospheric deposition. Transformations between labile and recalcitrant forms of N are generally biologically-driven with N lost to the atmosphere (through denitrification and volatilization) or to depth as soils drain (Baulch et al., 2011, Madramootoo et al., 2007). P exists in soils in both organic and inorganic forms, the latter derived from weathering of apatite. Like N, P enters the landscape through fertilizers, plant residues, and atmospheric deposition but is generally regarded to be less available due to soil sorption processes that are dependent on factors such as pH, temperature, and organic carbon content (Holtan et al., 1988). Phosphorus can be lost in runoff water, especially when concentrations exceed the sorption capacity of the soil, or when particulate P is transported along with soil through erosion processes. Soil frost can increase nutrient export by decreasing infiltration hence increasing soil–water interactions at the surface where soil P concentrations are often the highest (Cade-Menun et al., 2013). Additionally, freeze–thaw cycles disrupt plant cells and increase nutrient leaching from residues and other vegetation (White, 1973, Liu et al., 2013a, Costa et al., 2019a, Liu et al., 2019), which can become an important additional source of nutrients during snowmelt, particularly in the presence of young and actively growing plants (Cober et al., 2018, Elliott, 2013). The impact of tillage practices on nutrient export is complex. Conservation tillage can cause the accumulation of plant residue on farm fields, which can release nutrients to snowmelt runoff (Timmons et al., 1970, Miller et al., 1994, Ulén, 1997). In addition, by decreasing the mixing of the applied fertilizer, reduced tillage increases nutrient soil stratification and can lead to higher nutrient concentrations in surficial soils, which can be readily mobilized by runoff.

More reliable predictions of nutrient transport in cold agricultural basins have long been seen as a crucial to support nutrient management in Canada (Costa et al., 2020a, Baulch et al., 2019, Costa et al., 2019b). (Mekonnen, 2016) identified 74 models of water quality worldwide, but it has been noted that application of many of these models can be problematic in cold climates due to inadequate representation of many cold regions processes (Han et al., 2010). Costa et al. (2020a) reviewed the suitability of five prominent catchment nutrient models for application in cold climates: SWAT (Arnold et al., 1998), INCA (Whitehead et al., 1998, Wade et al., 2002), HYPE (Lindström et al., 2010, Arheimer et al., 2012), HSPF (Bicknell et al., 1997, Bicknell et al., 2005, Duda et al., 2012), and AnnAGNPS (Bosch et al., 1998). They identified inadequate representation of cold climate hydrology and daily time steps to be some of the features most commonly limiting the utility of these models in cold regions. They noted that most of these models have rarely been applied to cold regions, with the exception of SWAT and HYPE. They also found that some models allowed limited soil vertical resolution (i.e., maximum number of soil layers) that could reduce their performance in heavily stratified soils. Erosion remains a major challenge and meaningful model structures based on observable and transferable parameters were recommended to reduce the often high number of parameters for controlling biogeochemical transformations (leading to parameter identifiability). It was also highlighted that representations of accumulation of immobile nitrogen and phosphorus organic pools were often limited in their ability to represent legacy N and P for long-term simulations.

The meteorological data typically used to force hydrological models (e.g. solar radiation, air temperature, precipitation and wind speed) are often measured on a daily basis. This may limit the temporal resolution of model simulations and compromise their ability to capture hydrological and transport-biogeochemical processes that may be subject to significant diurnal variations (e.g. wind redistribution of snow and radiation-driven snowmelt) and episodic oscillations (e.g., sediment erosion and soil nutrient release). Unfortunately, in cold regions, hydro-biogeochemical processes activated during spring snowmelt and convective storms are often responsible for most of the annual nutrient export (Baulch et al., 2019, Kokulan et al., 2019). Thus, long-term simulations must also capture short-term runoff events, meaning that daily timestep models are insufficient. For example, the HSPF model is one of the few nutrient models identified that are often run at (default) hourly time intervals for long-term simulations. However, like many other models, HSPF does not explicitly account for some critical cold regions processes such as blowing snow and infiltration to and runoff over frozen soils, and uses the daily time step empirical degree-day method to estimate snowmelt, a method that has long been found to be inadequate in many cold regions (Walter et al., 2005, Gray and Landine, 1988).

There is a need to investigate alternative modelling approaches that are more applicable to cold agricultural basins, better reflecting cold regions hydrological and biogeochemical processes, which have a crucial impact on timing, concentration and load of nutrients. For this purpose, a complex of new hydro-biogeochemical modules was developed for the flexible, modular Cold Regions Hydrological Modelling platform (CRHM). CRHM has been created specifically to improve the simulation of cold regions hydrology (Pomeroy et al., 2007) and has been applied successfully to agricultural basins with minimal or no calibration (Fang and Pomeroy, 2008, Fang et al., 2010, Mahmood et al., 2017, Cordeiro et al., 2017, Costa et al., 2017, Kokulan et al., 2019). Its merit as a flexible and fundamentally physically based cold regions hydrological model renders it an ideal model for incorporating nutrient processes, and hence to offer a more suitable modelling framework to support nutrient management in agricultural cold regions.

Section snippets

The Cold Regions Hydrological Model

The Cold Regions Hydrological Model (CRHM) has been developed from more than 55 years of research on Canadian hydrology (Pomeroy et al., 2007). It is a modular platform that discretizes the basin into hydrologically distinct landscape elements called Hydrological Response Units (HRUs). It provides a range of predictive methods embedded in various modules that can be selected depending on dominant climatic and regional settings, i.e., mountains and prairie environments and are applied to

Hydrology

Fig. 4 compares observed and simulated SWE in fields F3 and F4 (locations are shown in Fig. 3). The observations correspond to the average snow accumulation peaks measured at the onset of spring snowmelt. The results show that the model can capture both the interannual and spatial variabilities in SWE distribution. Substantial heterogeneity in annual snow accumulation can be noticed within each field - note the standard deviation (error bars) for each year as a measure of the spatial variation

Critical management of timing of fertilizer use relative to major runoff events

Spring snowmelt is frequently the major annual nutrient export event in the Canadian Prairies, but fertilizer and manure applications in the growing season can also be mobilized via summer and spring rainfall-runoff events (Nicholaichuk, 1967, Hansen et al., 2002, Glozier et al., 2006, Liu et al., 2013b). This model and model application show that the timing of nutrient applications plays a key role in the amount of nutrients exported via runoff in southern Manitoba. Fig. 10 highlights that by

Conclusions

A series of process-based transport and biogeochemical modules have been developed for the Cold Regions Hydrological Model (CRHM) to simulate nitrogen (N) and phosphorus (P) in cold agricultural basins. The new model aims to address critical issues with existing nutrient models for simulation of these environments.

The new modules calculate nutrient fluxes throughout the basins’ hydrological system that includes the snowpack, soil, streams and depressional storage (e.g., potholes and wetlands).

Data Availability Statement

The data that support the findings of this study are available from Agriculture and Agri-Food Canada (AAFC) and Environment and Climate Change Canada (ECCC). Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the authors with permission of AAFC and ECCC.

CRediT authorship contribution statement

Diogo Costa: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. John W. Pomeroy: Conceptualization, Formal analysis, Funding acquisition, Supervision, Software, Writing – review & editing. Tom Brown: Formal analysis, Software. Helen Baulch: Supervision, Writing – review & editing. Jane Elliott: Supervision, Writing – review &

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This research was supported by the Global Water Futures Program and Environmental and Climate Change Canada. The authors would like to thank Agriculture and Agri-Food Canada and Environment and Climate Change Canada for kindly providing the data used in this study. The research would not have been possible without the interest and cooperation of landowners in the South Tobacco Creek Basin and the Deerwood Soil and Water Management Association.

References (100)

  • P.G. Whitehead et al.

    A semi-distributed integrated flow and nitrogen model for multiple source assessment in catchments (INCA): Part II - Application to large river basins in south Wales and eastern England

    Science of the Total Environment

    (1998)
  • M.R. Williams et al.

    Uncertainty in nutrient loads from tile-drained landscapes: Effect of sampling frequency, calculation algorithm, and compositing strategy

    Journal of Hydrology

    (2015)
  • D. Zhang et al.

    SWAT-CSenm: Enhancing SWAT nitrate module for a Canadian Shield catchment

    Science of the Total Environment

    (2016)
  • B. Arheimer et al.

    Water and nutrient simulations using the HYPE model for Sweden vs. the Baltic Sea basin – influence of input-data quality and scale

    Hydrology Research

    (2012)
  • J.G. Arnold et al.

    Large area hydrologic modelling and assessment. Part I: Model development. JAWRA

    Journal of the American Water Resources Association

    (1998)
  • H.M. Baulch et al.

    Nitrogen enrichment and the emission of nitrous oxide from streams

    Global Biogeochemical Cycles

    (2011)
  • H.M. Baulch et al.

    Soil and water management practices: Opportunities to mitigate nutrient losses to surface waters in the Northern Great Plains

    Environmental Reviews

    (2019)
  • B.R. Bicknell et al.

    Hydrological Simulation Program-Fortran, User’s manual for version 11: U.S. Environmental Protection Agency

    National Exposure Research Laboratory, Athens, Ga.

    (1997)
  • B.R. Bicknell et al.

    Hydrological Simulation Program – Fortran: HSPF Version 12.2 User’s Manual

    Technical Report

    (2005)
  • F. Birgand et al.

    Nitrogen Removal in Streams of Agricultural Catchments–A Literature Review

    Critical Reviews in Environmental Science and Technology

    (2007)
  • Bosch, D., Theurer, F., Bingner, R., Felton, G., 1998. Evaluation of the AnnAGNPS water quality model. Agricultural...
  • B.J. Cade-Menun et al.

    Nutrient loss from Saskatchewan cropland and pasture in spring snowmelt runoff

    Canadian Journal of Soil Science

    (2013)
  • C. Campbell et al.

    Mineralization rate constants and their use for estimating nitrogen mineralization in some Canadian prairie soils

    Canadian Journal of Soil Science

    (1984)
  • C.O. Clark

    Storage and the unit hydrograph

    Transactions of the American Society of Civil Engineers

    (1945)
  • J.R. Cober et al.

    Nutrient Release from Living and Terminated Cover Crops Under Variable Freeze-Thaw Cycles

    Agronomy Journal

    (2018)
  • Cordeiro, M.R.C., Wilson, H.F., Vanrobaeys, J., Pomeroy, J.W., Fang, X., Team, T.R.-A.P.B.M., 2017. Simulating...
  • D. Costa et al.

    A modelling framework to simulate field-scale nitrate response and transport during snowmelt: The WINTRA model

    Hydrological Processes

    (2017)
  • D. Costa et al.

    The temporal dynamics of snowmelt nutrient release from snow-plant residue mixtures: an experimental analysis and mathematical model development

    (2019)
  • D. Costa et al.

    Using an inverse modelling approach with equifinality control to investigate the dominant controls on snowmelt nutrient export

    Hydrological Processes

    (2019)
  • D. Costa et al.

    Preferential elution of ionic solutes in melting snowpacks: improving process understanding through field observations and modelling

    Science of The Total Environment

    (2019)
  • Costa, D., A Sexstone, G., W Pomeroy, J., H Campbell, D., W Clow, D., and Mast, A., 2020a. Preferential elution of...
  • D. Costa et al.

    Predicting Variable Contributing Areas, Hydrological Connectivity, and Solute Transport Pathways for a Canadian Prairie Basin

    Water Resources Research

    (2020)
  • W. Crumpton et al.

    Fate of non-point source nitrate loads in freshwater wetlands: results from experimental wetland mesocosms

  • Davies, T.D., Brimblecombe, P., Tranter, M., Tsiouris, S., Vincent, C.E., Abrahams, P., Blackwood, I.L., 1987. The...
  • J. Deelstra et al.

    Runoff and nutrient losses during winter periods in cold climates–requirements to nutrient simulation models

    Journal of Environmental Monitoring

    (2009)
  • P.F. Dornes et al.

    Influence of landscape aggregation in modelling snow-cover ablation and snowmelt runoff in a sub-arctic mountainous environment

    Hydrological Sciences Journal

    (2008)
  • Duda, P., Hummel, P., Jr, A.D., 2012. BASINS/HSPF: Model use, calibration, and validation. Transactions of...
  • E.W. Duncan et al.

    Linking Soil Phosphorus to Dissolved Phosphorus Losses in the Midwest

    Agricultural & Environmental Letters

    (2017)
  • J. Elliott

    Evaluating the potential contribution of vegetation as a nutrient source in snowmelt runoff

    Canadian Journal of Soil Science

    (2013)
  • J. Elliott et al.

    Influence of tillage and cropping system on soil organic matter, structure and infiltration in a rolling landscape

    Canadian Journal of Soil Science

    (1999)
  • R. Essery et al.

    Implications of spatial distributions of snow mass and melt rate for snow-cover depletion: theoretical considerations

    Annals of Glaciology

    (2004)
  • X. Fang et al.

    Drought impacts on Canadian prairie wetland snow hydrology

    Hydrological Processes

    (2008)
  • X. Fang et al.

    Prediction of snowmelt derived streamflow in a wetland dominated prairie basin

    Hydrology and Earth System Sciences

    (2010)
  • N.E. Glozier et al.

    Water quality characteristics and trends in a small agricultural watershed: South Tobacco Creek, Manitoba, 1992–2001

    (2006)
  • R.J. Granger et al.

    Snowmelt infiltration to frozen Prairie soils

    Canadian Journal of Earth Sciences

    (1984)
  • K.N. Grant et al.

    Differences in preferential flow with antecedent moisture conditions and soil texture: Implications for subsurface P transport

    Hydrological Processes

    (2019)
  • D.M. Gray et al.

    An energy-budget snowmelt model for the Canadian Prairies

    Canadian Journal of Earth Sciences

    (1988)
  • D. Gray et al.

    Densities of prairies snowpacks

  • D. Gray et al.

    Modelling Snowmelt Infiltration and Runoff in a Prairie Environment

  • D.M. Gray et al.

    Estimating areal snowmelt infiltration into frozen soils

    Hydrological Processes

    (2001)
  • Cited by (0)

    View full text