When rain washes soil from farmland into rivers, lakes, and streams, agrochemicals that can harm aquatic wildlife and raise water treatment costs, are often transported along with the silt particles. This sort of pollution can be reduced – but only if you know where it is coming from.
The pollution of freshwaters by agriculture is an increasingly challenging problem the world over. Whilst the conditions on farms that lead to pollution – and their solutions – are generally well understood, pinpointing the specific land uses or soil types that are the sources of silt reaching freshwater is much harder – especially when you are talking about fields lining hundreds of miles of riverbanks and ditches, each of which could lose lots of soil, no soil at all, or something in between.
Just how much of a problem such ‘diffuse pollution’ is regarded by governments is most clearly demonstrated by the eye watering sums of money thrown at it each year. An increasingly common way to tackle the problem has been through agri-environment initiatives as a means of providing advice and grants to farmers to help them address the issue themselves.
Between 2007 and 2013 over €22 billion of Common Agricultural Policy payments in the EU were delivered through such schemes. Similarly, the Conservation Security Program in the USA delivered $8 billion in grants between 2009-2018 to undertake and maintain conservation measures on farms.
The advice delivered relates to improving farming practices such as, when and how to plough to reduce erosion, how to improve soil structure, the management of farm tracks to reduce runoff down them, the management of livestock grazing, and the proper use of pesticides, fertilisers and slurry/manure. All of these should help limit the loss of soil and any associated pollutants being washed from the farm into freshwaters.
Whilst this is on the face of it a reasoanble approach, it doesn’t identify and target the actual sources of the pollution, but rather attempts to limit the most likely sources of it. In fact, the problem with this approach is that it likely isn’t delivering a sufficient improvement to meet policy targets given the significant monetary investments made.
Take the UK, where only about 16% of its freshwaters are in ‘good ecological status’. Here the Catchment Sensitive Farming initiative was set up to address this and has so far engaged with 19,776 farms covering 34% of England and has recently been expanded to cover the whole of the country with an annual budget increased from £15m to £30m.
However, when myself and my colleague Prof. Adrian Collins looked at eight priority river catchments targeted by this initiative, we found that whilst the advice being delivered was generally good, in some catchments there were mismatches between the opinions of Catchment Sensitive Farming officers as to the most important pollutant sources to target, and the actual sources responsible for the sediment problems in most of the catchments studied.
These findings highlight an increasing need for accessible and reliable information on sediment sources with which to target advice and grants. Therefore, a scientific method that has recently seen greater adoption worldwide is what is termed ‘sediment source fingerprinting’. This method compares the properties of the silt, or sediment as it is termed scientifically, collected from a stream to those of its potential sources present in its catchment area. These properties might include things like the concentration of elements such as iron, the activities of different radioactive isotopes, or the presence of specific organic molecules. Using this technique, a percentage contribution from each source to the silt collected from a stream can be estimated.
‘Fingerprinting’ as a Scientific Tool
Despite a long-standing aspiration to transform the fingerprinting approach from a specialist scientific tool into one being routinely used for catchment management at scale, it largely remains limited to use in academic contexts. In some cases, catchment managers or agencies may commission researchers to perform a sediment source fingerprinting investigation on their behalf – but the costs associated with such contracts are high due to the expertise required, time consuming sample collection and preparation, and expensive laboratory measurements of soil and sediment properties.
However, some of the fundamental components of a typical sediment fingerprinting procedure are simple and could be used with only a basic ability to use spreadsheets and create graphs. As a result, Prof. Collins and I have recently been developing a low cost, relatively low-tech way to help track down the sources of water pollution from farmland.
The idea builds on the rationale behind the tracing of sediment sources using inexpensive visible or infra-red spectra to match pollutant samples to their possible sources. This method is being increasingly used but still requires specialist laboratory equipment. Instead, we thought, why not use an ordinary office document scanner to do much the same job?
We set about comparing the use of the scanner with a more conventional fingerprinting approach to determine if this simple method is reliable. We wanted to look at this from the standpoint of a catchment manager or advisor, aiming to use the fingerprinting results to target mitigation of catchment sediment losses from the most important sources, and therefore, delivering optimum value for the public money invested and the efforts of farmers.
“in two study catchments, sediment colour was just as effective as conventional sediment fingerprinting”
We collected sediment and soil samples from eight river catchments. Once sediment and the samples of its possible sources had been dried and sieved to a consistent particle size, we put them in clear sample bags and placed them on a standard office document scanner. The captured image was then analysed using freely available software to obtain average red, blue and green values for each sample, which were then plotted against each other in simple graphs.
And the Results?
Well, in two study catchments, sediment colour was just as effective as conventional sediment fingerprinting. In four of the catchments,
sediment colour provided less source discrimination or minor differences in results, but still identified the most important targets for sediment mitigation. In only two catchments was colour unsuccessful; however, significant challenges were also encountered for these catchments even when using more conventional sediment fingerprinting methods due to specific local conditions.
So, our findings suggest that use of low-cost colour tracers by non-experts does have the potential to significantly increase the uptake of sediment source tracing. This would no doubt improve the cost-benefit of agri-environment initiatives when attempting to combat the degradation of water quality and aquatic ecology by excess sediment losses.
Adapting the Methodology
Next, we wanted to look at why the use of colour didn’t work in the two stream catchments mentioned above and improve the reliability of the method. We therefore looked at adding in a step where hydrogen peroxide was added to samples to remove organic matter and potentially improve source discrimination based solely upon geology or soil type. It was typically found in the previous catchments studied, that sediment colour was primarily controlled by land use with woodland or grassland soils having a darker colour than cultivated soils due to their higher organic matter content. This masked any effects of different soil types on its colour and limited how precisely we could identify critical sediment source areas.
The River Avon in southwest England and Holbeck/Wath Beck in northeast England were used to trial this revised method as they have both been identified as being of high priority for targeted management through a large agri-environment initiative.
The studied part of the River Avon catchment has an area of 28.5 km2 and is positioned in a small area of land surrounded by the uplands of the North Wessex Downs to the north and Salisbury plain to the south. The chalky hillslopes of the Downs were mostly covered in large fields growing maize and cereals, whilst in the valley bottom, peaty areas generally had smaller grassy fields, some with livestock.
A major question regarding this site was, if the cultivated land over the sloped chalk geology was the dominant source of sediment due to its higher potential erosion risk? Typically, erosion rates on cultivated fields are significantly higher than on grassland and steep slopes are associated with increased erosion. Therefore, this source might be expected to dominate catchment sediment losses.
But the tracing results showed that this high-risk source was not contributing significantly to the sampled sediment. This finding can be explained by the fact that the chalk hills are separated from most drainage channels by flat land, and sediment eroded from these hills lack pathways through which to reach watercourses.
With the most visibly high-risk fields discounted as a source of fine sediment, the flat valley bottom land over the greensand and peat geologies was identified as the most important sediment source. As this area is mostly used as grassland, most of which is not being grazed presently, options for mitigating sediment losses here are limited.
The combined catchment of Holbeck and Wath Beck (area 105 km2) is to the south of the North Yorkshire Moors and contains part of the Upland Howardian Hills. The catchment is characterised by a flat valley floor in the north which covers approximately a third of the catchment. The valley floor is surrounded by hills which are most extensive in the south and cover the remaining area.
Most of the catchment is cultivated for wheat, barley, oats, potatoes, rape and field beans, with a small number of fields used as grassland for sheep grazing or hay production. There is an outdoor pig farm located in the southeast of the catchment. Large patches of commercial pine forestry are present, with smaller areas of deciduous woodland.
Despite approximately 60% of this catchment being covered by sandstone hills which have a sandy soil texture, are predominantly cultivated and which therefore represent a high erosion risk, sediment contributions from this area were indicated to be low. As with the chalk hills in the River Avon, this finding can be explained firstly by a low stream density and absence of artificial draining ditches. The stream channels here are also often bordered by low erosion risk woodland or grassland fields meaning that few cultivated fields are immediately adjacent to watercourses – despite them covering much of the sandstone hills area.
The density of stream channels on the flat valley floor – which most sediment is indicated to originate from – is significantly higher, and some artificial ditches are present, meaning that more fields are immediately adjacent to watercourses. Additionally, woodland and grassland are much less common here, resulting in a higher proportion of the stream banks being adjacent to cultivated fields.
Despite some big differences in land use and geology between the two sites investigated, sediment colour was effective at identifying that a small proportion of each catchment which would normally be classified as low erosion risk, was in fact the dominant source of sediment. The advice to farmers and the conventional risk-based approach behind it, in the absence of this sediment source data, would almost certainly have got it wrong in both these cases.
The hydrogen peroxide sample treatment was able to improve the use of colour for sediment tracing in two ways when compared to using colour only with untreated samples. Firstly, sediment sources were able to be more precisely spatially refined due to the removal of organic matter, meaning that sediment colour reflected soil type rather than land use, which is the most significant control on colour without the treatment. Secondly, any alterations to sediment colour caused when it was washed across the landscape due to its enrichment in organic matter were removed.
This new work clearly demonstrates the challenges associated with predicting which sediment sources dominate catchment scale losses to water based on perceived erosion risk. In both catchments, the presence of pathways through which eroded soil could be transported to a ditch or stream was indicated to be the major control on sediment source. Unfortunately, it is difficult to identify these subtle and often diffuse pathways through field-based observations within a landscape or through mapped data.
We believe that this new sediment tracing method can potentially deliver accessible, inexpensive, precise, and reliable information to catchment managers at scale, leading to improved environmental outcomes from the expenditure of the public purse.