Skip to content

Future Landscapes

Sources and Flows

Managing contaminant pathways and attenuation to create headroom for productive land use

Saf Schematic

What Are We Doing?

About half (55%) of agricultural contaminants (nitrogen, phosphorus, sediment, faecal microbes) are naturally removed as they flow from land to sea – but this percentage varies greatly for different soils and landscapes.

Land-owners need site-specific information about the sources of these contaminants, their rate of transport to receiving waters, and how they are diluted while transported. This knowledge will enable the identification of critical source areas to target cost-effective mitigation interventions, and contribute to our understanding of the potential effects of different land uses on the environment (land use suitability).

The Sources and Flows programme brought together climate, soil, hydrology and water quality scientists to develop a multi-component framework that integrates and fills key gaps in knowledge about the source, transport and transformation of agricultural contaminants. The components of this framework can be incorporated into existing water quality tools and used in the development of new tools.

Sources and Flows researchers also assessed the effectiveness of efforts over the past 20 years to mitigate the impacts of farming on water quality.

 

How Can The Research Be Used?

  • The multi-component, modular framework developed by the Sources and Flows programme is a screening tool for quick assessment of water quality at catchment scale. It will allow landowners to pinpoint a location, identify the contaminants of concern, and select appropriate mitigations.
  • The mitigation assessments done by Sources and Flows researchers will help assess whether water quality objectives can be achieved in water bodies, or whether land use change might be required in the catchment area.
  • This research refined a ‘critical source area’ theory that explains that the majority of contaminants come from a minority of a farm or catchment. When targeted to these small but critical areas, mitigations are 6 to 7 times more cost-effective. This theory is now used in 77 guidelines, industry strategies and policy documents for environmental farm plans to improve water quality.
  • Sources and Flows researchers modelled soil drainage at 3 North Island and 3 South Island sites and found 35% to 80% more drainage peaks than expected. This suggests better monitoring of soil moisture and scheduling of irrigation is required. Because of this research the Northland Regional Council is refining drought intensity, duration and frequency curves for Northland to guide regional irrigation schemes and the allocation of water.
  • Sources and Flows research used precision irrigation technologies, timed and matched to soil type at a location in Central Otago, to show that the loss of nitrogen and phosphorus in soil drainage was reduced by 70% to 80% compared to standard irrigation practice. Irrigation NZ highlighted this nationally.
  • The Sources and Flows team is working with DairyNZ to develop geospatial data layers that capture the transport of N, P and faecal microbes from sources to water bodies. DairyNZ has been providing the research team with key data layer support as well as research questions that need to be answered.
  • Researchers in this programme created a map of New Zealand's potential groundwater recharge zones, where water moves downward from the surface to become groundwater. The map can be used as an initial guide for nationwide assessment and management of groundwater resources, and to identify areas of high nutrient leaching in zones with high groundwater recharge potential.
  • The Sources and Flows team has provided lookup tables of N and P losses to support the assessment tool being developed by Land Use Suitability research. This research will also guide the development of catchment modelling tools.

In the Media

Precision irrigation helps reduce nutrient losses

Irrigation NZ News, Autumn 2018

IrrigationNZ Chief Executive Andrew Curtis says the research is exciting and could be important in helping farmers reduce their nitrogen and phosphorus losses

VIEW ARTICLE →
Explainer: New “Swimmable” Water Standards

NZ Geographic

McDowell thinks the new standards are “… a substantial advance on what we’ve got at the moment”. The supplementary question is: are they good enough to meet everybody’s expectations, everywhere? And of course, the answer is no

VIEW ARTICLE →

Community Involvement

  • Greater Wellington Regional Council, Waikato Regional Council, Tasman Regional Council and Environment Southland have collaborated with the Sources and Flows programme to test the accuracy of national-scale mapping of groundwater redox in their regions. ECan chief scientist Dr Tim Davie participated in the initial stakeholder advisor group.
  • Environment Southland is a partner in developing and testing the Sources and Flows framework in the Oreti and Aparima catchments.
  • Sources and Flows work has also been tested in the Waiotapu (Waikato Regional Council ) and Waitangai (Northland Regional Council) catchments.
  • Sources and Flows has worked with Northland Regional Council to apply the drought risk component of their framework as a tool for the Northland region.
  • Sources and Flows had a stakeholder advisory group to guide initial dialogue with industry groups including Pamu, Beef + Lamb NZ, DairyNZ and Rabobank regarding their needs from the Sources and Flows framework.

Team Snapshot

Research Outputs

PAPERS

Transferring the impacts of pilot-scale studies to other scales: Understanding the role of non-biophysical factors using field-based irrigation studies

Graeme Nicholas, MS Srinivasan, Sam Beechener, Jeff Foote, Melissa Robson-Williams, Stephen FitzHerbert
Agricultural Water Management, April 2020

Researchers are challenged to design research that can generate credible claims regarding cross-scale impact and adoption. However, the context in which new knowledge or innovation is developed and tested may differ from that for the uptake and use of those findings. This paper reports insight into the problem of designing impactful research and proposes a model to assist bio-physical researchers in accounting for non-biophysical context when moving between scales or settings. We treat the scaling problem as the one shifting contexts. The use of the model is illustrated by application in two New Zealand-based irrigation water use efficiency (WUE) field studies. We hypothesised that to successfully transfer the learnings from these two pilot studies to other scales it would be important to understand the influence of context on WUE practices. To support this process, we developed a social dynamics model (Composite Context Model, CCM) from existing social systems frameworks. The CCM maps influential non-biophysical dynamics to help interpret the WUE field study findings for other scales. The paper represents a contribution to researchers addressing two related challenges: that of making credible claims regarding the possible future impact of their research, and that of translating innovations across scales. By demonstrating the use of our CCM for documenting key non-biophysical variables, we aim to equip researchers with a practical tool to assist in the interpretation of findings across contexts, that include both biophysical and non-biophysical factors.

Achieving unbiased predictions of national-scale groundwater redox conditions via data oversampling and statistical learning

S. Wilson, et al.
Science of the Total Environment, 2020

An important policy consideration for integrated land and water management is to understand the spatial distribution of nitrate attenuation in the groundwater system, for which redox condition is the key indicator. This paper proposes a methodology to accommodate the computational demands of large datasets, and presents national-scale predictions of groundwater redox class for New Zealand. Our approach applies statistical learning methods to predict redox status in areas without sample data. A key achievement was to overcome the influence of sample selection bias on model training via oversampling. National maps are provided for redox class and probability at specified depths. Our model provides unbiased predictions at a scale relevant for environmental policy, and enables targeted interventions that can achieve the desired environmental outcome in a more cost-effective manner than non-targeted interventions.

Comparison of four learning-based methods for predicting groundwater redox status

Friedel et al.
Journal of Hydrology, 2020

Knowing the location where groundwater denitrification occurs, or by proxy the groundwater redox status (oxic, mixed, and anoxic), is valuable information for assessing and managing potential agricultural land-use impacts on freshwater quality. We compare the efficacy of supervised and unsupervised learning-based methods to predict groundwater redox status in the agriculturally dominated Tasman, Waikato, and Wellington regions of New Zealand. Overall, the supervised methods demonstrate a prediction bias toward oxic conditions and inability to perform statistically well when using independent regional data. By contrast, the unsupervised method performs statistically well when predicting oxic, mixed, and anoxic conditions and corresponding depths when using independent regional data. The unsupervised learning method provides added benefits.

Development of a national-scale framework to characterise transfers of N, P and Escherichia coli from land to water

M. S. Srinivasan, Richard W. Muirhead, Shailesh K. Singh, Ross M. Monaghan, Roland Stenger, Murray E. Close, Andrew Manderson, John J. Drewry, Leo Christopher Smith, Diana Selbie, Roger Hodson
New Zealand Journal of Agricultural Research, January 2020

A hydrological framework encompassing nitrogen (N), phosphorus (P) and microbial (E. coli) transfer from land to water was developed to provide a consistent and rapid approach for assessing the potential impacts of land activity on water quality in New Zealand. Application of the framework in four catchments highlighted the importance of local catchment knowledge of dominant hydrological processes that was needed to ensure flow partitions derived were a realistic representation of transport processes. Additional refinements are needed to improve process representation (e.g. effects of groundwater lags) and ensure input data (e.g. soil attributes) have appropriate resolution to describe hydrological pathways. We contend that such a framework would provide a consistent and relatively rapid approach for identifying contaminant transfer pathways from land to water that can inform assessments of the potential consequences of land use change and intensification.

The effectiveness of streambank fencing to improve microbial water quality: a review

R. Muirhead
Agricultural Water Management, August 2019

This literature review collates published data on the effectiveness of fencing stock out of waterways to reduce faecal indicator bacteria concentrations in streams. Eighteen suitable papers were identified, from 4 countries. Reported values of stream fencing effectiveness ranged from zero to 96%. Overall, different experimental approaches did not appear to bias results. This large range in effectiveness values indicates that while we are confident that stream fencing will improve microbial water quality in streams, we have very low confidence in predicting the actual percentage reduction values that can be expected.

CLUES model calibration: residual analysis to investigate potential sources of model error

A. Semadeni-Davies, C. Jones-Todd, M.S. Srinivasan, R. Muirhead, A. Elliott, U. Shankar, C. Tanner
NZ Journal of Agricultural Research, 2019

This paper presents the results of the latest calibrations of the CLUES model for each of total nitrogen, total phosphorus and Escherichia coli. CLUES has been calibrated nationally against loads estimated from data collected from water quality monitoring stations located across the country which could lead to parameter bias. To ascertain whether there is any systematic bias, the calibration residuals were evaluated against regional (geographical) location and a range of upstream catchment characteristics. We found that CLUES gives reasonable load estimates at the catchment. However, there was significant uncertainty in the parameterisation. While several significant relationships were found between upstream catchment characteristics and the model residuals, these relationships were weak and are unlikely to point to any systematic bias in the calibration.

CLUES model calibration and its implications for estimating contaminant attenuation

A. Semadeni-Davies., C. Jones-Todd, M.S. Srinivasan, R. Muirhead, U. Shankar, C. Tanner, A. Elliott.
Agricultural Water Management, 2019

Catchment water quality models are essential for freshwater management; however, their value is dependent on their performance and level of uncertainty. Here we present the latest calibration of the Catchment Landuse for Environmental Sustainability model (CLUES) and examine the implications of the calibration for estimating in-stream contaminant attenuation. CLUES estimates catchment mean annual loads of Total Nitrogen, Total Phosphorus and E. coli and is widely used in New Zealand for both catchment planning and policy development. We found that CLUES gives reasonable load estimates at the catchment scale. However, there was significant uncertainty in the SPARROW parameterisation. We conclude that while CLUES can be used to estimate catchment contaminant loads, it cannot adequately estimate instream attenuation. Improved representation of lower order streams in the calibration data would allow us to evaluate the source yields from each of the model components to better estimate attenuation.

Human Impacts Recorded in Chemical and Isotopic Fingerprints of Soils from Dunedin City, New Zealand

R. Turnbull, K. Rogers, A. Martin, M. Rattenbuy, R. Morgan
Science of The Total Environment, July 2019

We present results from the first urban chemical and isotopic soil baseline survey to be completed for a New Zealand city. The major, minor, trace and isotopic composition of soils from different depths across the city of Dunedin are shown to be spatially variable due to geogenic and anthropogenic influences.

Grid‐based sediment tracing approach to determine sediment sources

A. Haddadchi, D. Murray Hicks, J. M. Olley, S. Singh, M.S. Srinivasan
Land Degradation and Development, July 2019

This grid‐based sediment tracing technique improves the precision of source contribution estimates and enhances the granularity of sediment source maps. We test the proposed technique using source and suspended sediment samples collected from the Emu Creek Catchment (911 km2), south‐east Queensland, Australia. Geochemical tracers were employed to distinguish sediments derived from the heterogenous and complex underlying rock types. Importantly, the proposed technique provided a greater spatial resolution of the sediment sources by assigning sediment contributions into grid sources rather than the area‐specific source types.

Potential groundwater recharge zones across New Zealand

S. Singh, M. Zeddies, U. Shankar, G.Griffiths
Geoscience Frontiers, May 2019

Knowledge of groundwater recharge potential is required for sustainable groundwater management, including the assessment of vulnerability to contamination. This study maps (at 500m x 500m resolution) the potential groundwater recharge zones across New Zealand using national data sets of lithology, slope, aspect, land use, soil drainage and drainage density. Weights to the factor was calculated by three different methods. Five potential groundwater recharge zones were defined across New Zealand. The maps can be used to identify areas of high nutrient leaching in zones where high groundwater recharge potential exists (regions with large lakes and in the lower elevation plains).

Towards baseflow index characterisation at national scale in New Zealand

Singh, SK.; Pahlow M.; Booker, DJ; Shankar, U.; Chamorro, A.
Journal of Hydrology, January 2019

BaseFlow Index (BFI) was determined for all river reaches in New Zealand. A recursive digital filtering technique was applied to 482 gauged sites across New Zealand, then an individual filter parameter was determined for each catchment. BFI varies between 0.20 and 0.96 with an average of 0.53, which indicates 53% of long-term streamflow in New Zealand is likely to originate from groundwater discharge and other delayed sources. This dataset can support water resources planning and management in New Zealand, in particular water supply, stream ecology and pollution risk. The methodology is applicable to any region around the world.

Indirect Methods to Elucidate Water Flows and Contaminant Transfer Pathways through Meso-scale catchments – a Review

Singh, SK; Stenger, R.
Environmental Processes, December 2018

We provide a critical review of indirect methods to elucidate water flows and contaminant transfer pathways through meso-scale catchments, as the proliferation of such methods in recent years has made it very difficult for potential users to evaluate their relative merits. Advantages and disadvantages in terms of data availability and underlying assumptions are highlighted to facilitate the selection of a suitable method.

Indirect faecal source tracking methods to elucidate critical sources and contaminant transfers through catchments – a review

Devane, M.L.; Weaver, L.; Singh, S.K.; Gilpin, B.J.
Journal of Environmental Management, May 2018

In New Zealand, there is substantial potential for microbial contaminants from agricultural fecal sources to be transported into waterways. Understanding contaminant transport pathways from catchment to stream can aid water management strategies. It is not practical to conduct direct field measurement for all catchments, so fecal source tracking can be utilised to link catchment characteristics to fecal signatures identifying critical sources. In this article, we have reviewed approaches to identifying critical sources and pathways for fecal microorganisms from agricultural sources, and make recommendations for the appropriate use of these fecal source tracking (FST) tools.

Variability of Escherichia coli Concentrations in Rivers during Base-Flow Conditions in New Zealand

Muirhead, R, Meenken, E
Journal of Environmental Quality, March 2018

We compared the variability of E. coli concentrations in baseflow of 3 different-sized rivers in both summer and winter at the time scales of minutes, hours, and days, to variability from laboratory replication of the measurement methods. Estimates of variability, analysed using the coefficient of variation (CV), were approximately 32% and 60% at hourly and daily time scales. There is strong evidence that both time scale (p < 0.001) and river (p < 0.001) significantly affect the variation in E. coli concentrations. This variability should be considered when interpreting the results from a one-off grab sample used to compare against water quality standards or for calibrating models.

Soil hydraulic modelling outcomes with four parameterization methods: comparing soil description and data inversion approaches

Graham, SL; Srinivasan, MS, Faulkner, N, Carrick, S.
Vadose Zone Journal, March 2018

Different methods for parameterizing soil hydraulic models can lead to substantially varied predictions of soil–plant–atmosphere water fluxes. This study investigated, for a heterogeneous stony soil, 4 methods of soil hydraulic parameterization. Soil drainage, volumetric water content, and evapotranspiration were modeled using HYDRUS-1D for an irrigated pasture in New Zealand. While all methods underpredicted evapotranspiration by 18% to 30% compared with eddy covariance, improvement in drainage estimates with inverse estimation from field data led to decreased capability for modeling evapotranspiration. We suggest this approach for application in other settings to select the most appropriate parameterization approach for a given soil hydraulic model application.

Does variable rate irrigation decrease nutrient leaching losses from grazed dairy farming?

McDowell RW
International Journal of Water Resources Development, December 2017

A six‐year study was conducted to determine whether the use of variable rate irrigation (VRI), compared to uniform rate irrigation (URI), could decrease N and P leaching losses from a 143‐ha area under intensively grazed dairy cattle that had been partly hydrologically isolated by the installation of artificial drainage pipes. After accounting for potential differences in flow, annual load estimates of N and P species at the downstream site under VRI were about 80–85% less than that lost under URI. Wider adoption of VRI technology could therefore decrease farm leaching losses and nutrient concentrations in receiving waterbodies.

Impacts of long-term biomass management on soil phosphorus under temperate grassland

Boitt G, Black A, Wakelin SA, McDowell RW, Condron LM
Plant and Soil - Plant Science, September 2017

We assessed and quantified the cumulative impact of 20 years of biomass management on the nature and bioavailability of soil phosphorus (P) accumulated from antecedent fertiliser inputs. Contemporary plant production and P uptake were over 2-fold higher for the biomass retained compared with the biomass removed regimes. Soil C, total P, soluble and labile forms of inorganic and organic soil P were significantly higher under biomass retention than removal.

Temperature and Nitrogen Effects on Phosphorus Uptake by Agricultural Stream-Bed Sediments

McDowell RW, Elkin, KR, Kleinman PJA
Jounral of Environmental Quality Abstract - Landscape and Watershed Processes, March 2017

Climate change will likely increase the growing season, temperatures, and ratio of nitrogen (N) to phosphorus (P) loss from land to water. However, it is unknown how these factors influence P concentrations in streams. We sought to evaluate differences in biotic and abiotic processes affecting stream sediment P dynamics under different temperature and N-enrichment regimes. Analysis of sediment samples indicated that P uptake via abiotic processes was greater at 19 than at 26°C. The addition of N stimulated P uptake by the microbial biomass at 19°C, but microbial uptake was potentially inhibited at 26°C. Because microbial biomass is a temporary store of P, these data suggest that more P may be available with increasing temperatures during the growing season, especially under baseflow, implying that strategies to mitigate P losses from land to water should be strengthened to prevent potential water quality impairment.

A copula-based analysis of severity-duration-frequency of droughts in six climatic regions of New Zealand

Singh, S.K; Chamorro, A; Srinivasan, M.S., Breuer, L
Journal of Hydrology, February 2017

We used a copula to model the joint drought duration-severity distribution and frequency to generate SDF curves for six locations in different climatic regions (3 in the North Island and 3 in the South Island). Based on a bivariate-fitted model for duration and severity, drought SDF curves for various recurrence intervals were derived. We found that North Island locations experienced more dry spells than the South Island locations. The SDF curves provide a comprehensive understanding of occurrence and duration of dry conditions, and can be a useful tool for natural resource managers in developing short and long-term drought mitigation strategies for water management and conservation.

REPORTS

Application of a revised SedNetNZ model to the Oreti and Aparima catchments, Southland

Hugh Smith, Alex Herzig, John Dymond, Les Basher
Report by Manaaki Whenua for OLW, June 2019

This report presents an application of a revised version of the SedNetNZ sediment budget model to the Oreti and Aparima catchments in Southland. We focused on modifying the surface and bank erosion components of the SedNetNZ model, because these widespread erosion processes are likely to dominate suspended sediment loads.

 

CONFERENCE PRESENTATIONS

Sources and Flows, Symposium 2019

Diana Selbie
Our Land and Water Symposium, August 2019

Watch video: https://vimeo.com/356108131

Investigation of methods to predict groundwater redox status using limited data

Close, M., Wilson, S., Friedel, M., Abraham, P., Banasiak, L.
NZ Freshwater Sciences Society Annual Conference, December 2018

CLUES calibration – can we use CLUES to estimate attenuation?

Annette Semadeni-Davies, Charlotte Jones-Todd, MS Srinivasan, Richard Muirhead, Sandy Elliott, Ude Shankar, Christopher Palliser, Chris Tanner
NZ Freshwater Sciences Society Annual Conference, December 2018

Mitigating the impacts of dairy farming on water quality - what have we achieved?

Monaghan R, Smith C, Manderson A, Muirhead R, Burger D, Eikaas H and R McDowell R
NZ Society of Soil Science, December 2018

Variability of E. coli in rivers during base-flow conditions

Richard Muirhead
Water Microbiology Conference, May 2018

Understanding the linkage between hydrological and chemical signatures at catchment outlets and dominant contaminant transfer pathways

Stenger, R.; Singh, S.; Muirhead, R.; Devane, M.; Basher, L.; Srinivasan, MS
Integrating Multiple Aquatic Values, November 2017

Investigation of methods to predict groundwater redox status with variable amounts of available well data

Murray Close, Scott Wilson, Mike Friedel, Phil Abraham, Laura Banasia
Hydrological Society Annual Conference, December 2018

Variability of E.coli in rivers: implications for interpretation of grab samples

Muirhead, R
NZ Freshwater Sciences Society Annual Conference, December 2018

Predicting NZ groundwater-redox status: machine-learning considerations & preliminary results

Friedel, M.J., Wilson, S., Close, M., Buscema, M., Abraham, P., Banasiak, L
Hydrological Society Annual Conference, December 2018

Linking water flow and contaminant transfer through meso-scale catchments

Singh, S.; Stenger, R.; Devane, M.; Basher, L.; Muirhead, R.; Srinivasan, MS
NZ Hydrological Society Conference, December 2017

Development of linked frameworks to represent and manage catchment-scale contaminant transport for improved water quality outcomes

Muirhead, R.; Srinivasan, MS.; Monaghan, R.; Basher, L.; Manderson, A.; Tipa, G
Freshwater Society Conference, November 2017

Have a Question?

We are happy to answer any questions about this research and how it can be used.

Please fill in the form below

Thank you for your enquiry. Your question may require information to be gathered from the research team, so please anticipate that it may take us up to 10 days to prepare a reply.

We appreciate your interest in our research.

  • This field is for validation purposes and should be left unchanged.
Scroll To Top