Having given some thought to why you are monitoring, what your objectives are and what impact you are aiming for, you’re ready to get into the nuts and bolts of monitoring – collecting your own data. In this section we’ll give you some pointers on the following:
- What data will you collect?
- What methods will you use to collect data?
- Where will you carry out your data collection?
- When and how often will you collect data?
- How will you ensure your data is of a known quality?
1. What data will you collect?
Rivers and river catchments are shaped by complex interactions between water, weather, land use, seasonal variations, plants, animals, biological, physical and chemical processes, soils, geology, people and infrastructure. When selecting which data to collect as part of a structured monitoring program you should consider a number of themes:
Biological
Aquatic invertebrates, fish, plants / wildlife sightings.
Aquatic organisms are in the water all the time so the number and variety of species in a location can provide a good indication of the health of the river. Other types of animals such as birds and mammals depend on invertebrates or fish as a food source, so the presence of otters, dippers or kingfishers can also be an important indicator of a healthy river habitat.
Methods: View biological methods (under development)
Water quality
Nutrients (e.g. nitrate, phosphate, ammonia), microbiology (e.g. E. Coli, faecal coliforms), temperature, conductivity, turbidity, dissolved oxygen, temperature, pH, chemicals
Often what people first think of when river monitoring, there are many different aspects of water quality that can be measured and many different methods available. Some parameters are easier to measure in the field than others. Measuring water quality – either by sampling or bank-side tests, gives a snapshot of conditions and water quality can change rapidly, particularly during rainstorms. It is usually important to gather a number of samples over a period of time (typically weeks or months) before drawing conclusions from that data – it is unwise to make assumptions from a single result.
Methods: View water chemistry methods (under development)
Physical condition or pressures
Habitat and hydromorphology, barriers
The term ‘hydromorphology’ refers to the physical character of a river and the dynamic processes that can lead to further degradation or recovery. The naturalness of a rivers form (i.e. degree of incision, type of substrate, naturally meandering or artificially straightened) will have a significant impact on the plants and animals that can live there and can be an indicator of wider issues in the catchment.
Methods: View physical habitat methods (under development)
Pollution pathways and/or issues
Non-native species, run-off pathways, litter/plastics or outfalls/discharges
Some pressures on a river are relatively easy to spot even though the level of impact might be hard to determine. For example, whilst we can be confident that road runoff or intermittent sewage discharges are harmful, quantifying the degree of negative impact can be very difficult, requiring multiple surveys or water quality tests/samples for some difficult parameters to measure (e.g. toxic metals or exotic chemicals) over a range of different conditions. In such cases it is usually more effective to record the visible (or odorous) impacts of known pollution sources – e.g. the presence of sewage fungus, sewage related litter and unpleasant smell is an indicator that something is wrong with a particular outfall pipe. A similar approach can be taken with active run off pathways (both urban and rural). Having a comprehensive record of the behaviour of these types of feature can help to speed up remedial action by the appropriate authority.
Methods: View testing protocols (under development)
Terrestrial catchment condition
Soil health, land use, hedgerows, tree cover etc.
Rivers are formed by water that falls on a wider catchment area and moves over the surface of the land or through the soil and rocks as groundwater. Therefore elements such as soil health, land use, artificial drainage and tree cover can have a significant impact on river health. If you are aiming to improve river health by encouraging or facilitating changes to the wider catchment then you may well consider structured monitoring of some of these elements.
Methods: View soil testing methods (under development)
A better understanding of river health is unlikely to result from monitoring any one of these themes in isolation, so whilst one of them may be identified as your main focus, consider collecting additional data from other themes whilst you are out and about. For example, you could take a photo from the same point each time you test for water quality or record wildlife sightings when carrying out invertebrate or morphology surveys. It can also be useful to record whether there has been recent rainfall as this can give important contextual information – high turbidity levels after heavy rainfall are commonplace, but during dry weather, it can be an indication of significant pollution.
2. What methods will you use to collect data?
Within each of the themes and data types outlined above there will be a range of specific methods to choose from. Which method(s) you choose will depend on a number of factors:
What methods are others using? If you are planning to share your data with others then it is worth considering alignment of methods. Sometimes this won’t be possible (e.g. where the Environment Agency or water company exclusively use laboratory analysis). It is also worth considering methods that are well used by other similar groups as there are more likely to be supporting resources and the data is more likely to be accepted by others.
What resources do you have available? Think about money, people, technical skills / experience and time.
Guidance: defining your primary monitoring purpose
What level of accuracy or robustness is required? Often there is a trade-off between quick, cheap and simple methods that give an indicative result and more expensive, technically demanding and/or time consuming techniques that give a more reliable or thorough result.
What are you (or others) going to use the data for? Whilst it can be tempting to assume that other people, groups or organisations will be eager to use your data many official bodies have become used to working with sparse data coverage. They may not have the resources to utilise new and large data sets. It is worth checking in with any key stakeholders as you are planning your monitoring to see what their data requirements might be.
Overview: View methods section
3. Where will you collect data?
The choice of monitoring or sampling locations will be strongly influenced by your monitoring purpose. Potential sample locations fall into two main categories depending on whether you are concerned about point source discharges or wider river/catchment health screening or assessment.
Point source discharges
You will want sample locations that are upstream and downstream of the source so that the impact of the discharge can be measured. Remember that effluent from a discharge will take a while to mix with the receiving waterbody – so the distance downstream will strongly affect the observed impact.
** It is rarely a good idea to take samples directly from a discharge pipe – this can be difficult, dangerous and unpleasant and is unlikely to yield useful data. There are other ways to assess the discharge from an outfall such as visual inspection, photographs and noting any unpleasant or distinctive odours. **
Catchment health screening/assessment
If your objective is to obtain a more general assessment of catchment health, then you will be looking for monitoring locations that are representative of the upstream catchment area – in this case, it is better to avoid locations that will be strongly influenced by a point source.
Regardless of the nature of the monitoring point, there are a number of factors that you need to consider.
Safe and legal access to the site
It is very important that a monitoring site can be reached safely and legally under a range of flow and weather conditions and ideally by people of varying physical abilities. Public areas such as parks can provide good locations, which may also be suitable for training sessions. Bridges make good monitoring points and many of the Environment Agency sampling points tend to be from bridges. Take care to ensure that there is a safe pedestrian walkway on the bridge from which a sample can be collected and avoid bridges on busy roads. You will also need to consider the height and robustness of any guard rail – is it too high to reach over? Or too low or insecure to be safe? Also, consider how your monitoring team will travel to the site – if they are driving, is there somewhere safe and legal to park a car nearby?
The confluence of two tributaries can make an efficient sampling location, as two sub-catchments can be sampled in one visit. In this case, make sure that the sample points are far enough upstream of the confluence, as you can get a ‘backwash’ effect where the flow from one tributary gets drawn some way up the other. As a guide try and find an access point that is 5 to 10 times the channel width upstream.
Access to the water
The type of access that is required will vary according to the type of surveying/sampling being done. Some surveys require a clear view of the river channel, some require the collection of a water sample and some require access to a stretch of river channel itself. Some sites will be suitable for all types of access, and some will not, so it is important to understand access restrictions, i.e. don’t assume that because water samples can be collected from a site, that it is automatically suitable for Riverfly or MoRPh surveys too.
- Visual surveys: Generally the least demanding type of access, but bear in mind that particularly in smaller channels vegetation can obscure much of the channel in summer
- Water sample collection: The level of access required to collect a water sample (either to be sent to a lab for analysis or for bankside testing) will depend on or dictate the sample collection equipment provided or available. Sampling from a bridge will usually require a small bucket or sampling can on a length of rope or line, this can also be suitable for sampling from a vertical bank. If the bank is sloping then a container on a pole or bamboo cane is usually more appropriate as the sample container should not be dragged across the ground surface or through vegetation. There are sites where it is acceptable to enter the water course directly to collect a sample – but this is usually a last resort. There are obvious risks associated with entering a river or stream channel and this will certainly not be possible during higher flows.
- Full access to the river channel and/or banks – some types of survey e.g. Riverfly or MoRPh surveys require full access to either the river banks or channel itself. These surveys will come with their own advice on safe access – but if planning a monitoring programme note that there will be a reduced number of sites where these types of survey will be practical, and that surveys are more likely to need to be carried out by pairs of volunteers.
Guidance: Health and safety
Number of locations
The number of sites that you include in your monitoring programme will depend on how many volunteers are monitoring and their appetite for spending time monitoring. It is important to consider the number of sites you will aim to cover alongside your initial objectives. If your coverage is too sparse, it will reduce the effectiveness of a catchment health screening programme. If you try and take on too many sites then you may overstretch your volunteer team and struggle to achieve consistent coverage of key sites.
The optimum spatial density of monitoring locations will be different for different monitoring ‘themes’. For water quality monitoring a site that is located towards the downstream end of a waterbody can be considered representative of the upstream catchment, although the degree to which issues from upstream will be detected at that site will decrease with distance. Ideally, and if resources allow, a series of monitoring points will be established throughout the stream network. As a general guide try to find suitable sampling locations around every 2 – 3 km of stream length. In larger rivers this can be increased to 5 – 10 km between sites and in towns and cities (where there tend to be more inputs from surface water drains or sewage infrastructure) the distance can be decreased to around 1 km. Try to identify a monitoring point on streams that are 1500 m or longer.
Detailed maps are invaluable for identifying potential monitoring locations. A paper copy of the Ordnance Survey map of your catchment would be a sound investment, the Explorer series at 1:25,000 scale will be most useful and you can buy custom maps that are centred on a location of your choice. There are also a number of online resources where different map products can be accessed. For example:
- Magic Maps – Defra mapping site that includes Ordnance Survey base maps, aerial imagery, and a host of other useful data layers
- Google Maps / Google Earth – online mapping including 3D, ‘Street View’ and historical imagery options
- Ordnance Survey online maps – basic level access is free, premium features available if subscribed
- Grid Reference Finder – a useful site for finding co-ordinates and converting between OS Grid references, lat/long, Eastings/Northings, What3Words etc.
- What3Words – online mapping with an accessible method of sharing a 3 m x 3 m location
- Catchment Data Explorer (England) – useful for identifying catchments and waterbodies and for seeing the latest Water Framework Directive assessment
- Environment Agency Water Quality Archive (England) – useful for identifying current and historic EA sampling locations (and water quality results)
Most County and/or District Councils have an online, interactive mapping service – these can be useful for identifying Public Rights of Way. Footpath or bridleway crossings of streams or rivers can make pleasant sampling locations – some Council mapping services (Somerset Council for example) even include photographs of all footpath or bridleway bridges crossing streams.
Using these resources, you can find out more about your catchment and identify potential monitoring points. Google Street View can also help identify ‘on the ground’ details, such as whether bridges are likely to make suitable sampling locations.
Scout out testing locations: Whilst you can find out a lot about potential monitoring points from the comfort of your own home, it is important that someone visits the site before including it in your monitoring network as the online information sources may be out of date, incomplete or may not convey the complete situation. Some schemes (such as Westcountry CSI) insist that volunteers visit any new site as a ‘dry run’ survey first, enabling them to evaluate the suitability of the location without any pressure to collect a water sample or carry out any tests.
Plan for volunteering levels to fluctuate: Ideally, volunteer monitoring programmes should be able to expand (or contract) with fluctuating numbers of volunteers, so having a network of potential sites identified and some means of prioritising which ones are visited is a good idea. For example, it may be that a key monitoring site is located towards the downstream extents of a sub-catchment – this will be one of the first that you assign a volunteer to. If more volunteers come forward, then you may choose to expand to another catchment or waterbody or include sites further upstream of your first monitoring site. If the volunteer that is assigned to one of your primary monitoring locations is away, loses interest of leaves the scheme, then ideally another volunteer would be recruited to continue at that site, or existing volunteers may be able to step in for a while.
Adapt to your volunteer needs: It can add to the interest level of volunteers if they regularly sample or survey more than one site, as they can compare and contrast what they find. Some volunteers may be keen to take on a large number of sites, especially when they first get involved. However, for most monitoring programmes, consistent data over the medium to long term is important, so it is usually preferable to make sure that volunteers commit to a level of involvement that is sustainable and enjoyable. It is usually beneficial if volunteers ‘adopt’ certain sites – familiarity with a site helps volunteers spot changes or abnormal conditions.
4. When will you collect data?
Rivers and river catchments are dynamic environments that are constantly changing. We have already looked at some of the elements that make up a river system (morphology, biology, habitat, water quality, etc.), and each of these can have a range of typical change periods – from minutes or hours with water quality to perhaps months or years with some habitat or morphological aspects.
Some elements will have periods during the day or time of year when conditions are less suitable for monitoring, perhaps because they are not representative of general conditions or because surveying may be difficult or unsafe. For example – all surveying/testing needs to be done in daylight hours; during the winter, this provides a much shorter window than in the summer; morphological surveys need good visibility of underlying physical features – this can be difficult in summer when vegetation is at its height.
Depending on the purpose of your monitoring programme, you may be planning to carry out monitoring indefinitely, or you may have a finite duration for monitoring in mind. If you have an open-ended monitoring plan, then you’ll need to consider what level of effort is sustainable over the medium to longer term, whereas if you are monitoring for a finite duration, then you’ll need to ensure you collect sufficient data to draw conclusions from.
The purpose of your monitoring will also influence the timing of your data collection – this is particularly relevant for water quality testing with many determinants being very different according to flow conditions (i.e. after heavy rainfall or a prolonged dry period), but it is also likely a consideration for other types of surveying/testing. If you are trying to establish baseline or prevailing conditions over a stretch of river or a catchment, then it is best to aim for a ‘regular but random’ monitoring schedule. This means that you would schedule monitoring at broadly regular intervals (likely weekly, fortnightly or monthly) but try and avoid performing tests at exactly the same time of the day or day of the week. If your monitoring objective is to establish the effect of specific conditions (e.g. heavy rainfall vs dry periods) then it makes sense to target your monitoring at those conditions.
Considering the monitoring themes that we set out previously, some suggested monitoring frequencies, durations and target times are set out below:
| Monitoring element | Frequency | Duration | Times to avoid / target |
| Biological (invertebrates, fish) | Monthly – Annually | Indefinite | Salmonid spawning periods, periods of very low flow |
| Water quality | Weekly – Monthly | Indefinite / year round | Extreme high flows due to safety concerns; try to vary times |
| Habitat | Bi-annually – Annually | Indefinite / seasonally dependant | Extreme high flows / peak vegetation |
| Pollution pathways | Ad hoc / event based | Indefinite | During rainfall / prolonged dry periods |
| Terrestrial | Seasonal / Annually | Indefinite | Species dependant |
Consider special event timing: Some pollution sources (wastewater treatment plants or industrial facilities) are known to have regular discharge cycles or times when effluent is of lower quality. So, if you always sampled at a site at 14:00 on a Saturday, you may always miss a pollution peak (thereby under-representing a problem), or your sampling may always coincide with a peak (thereby over-representing an issue). Due to the transit time of water flowing downstream, if you are some distance from a significant outfall, then those peak times can be several hours later. Many water quality parameters are subject to diurnal variations – particularly in warm summer conditions, due to a cycle of photosynthesis/respiration driven by higher temperatures and sunlight during the day. This is particularly pronounced with dissolved oxygen and temperature – which both tend to peak in the late afternoon and reach a low point just before dawn. Sampling at the same time each day risks disproportionately capturing data closer to one of those peaks.
In practice, it is usually easier for volunteers to incorporate regular monitoring into their routine, which may mean sampling happens at similar times of the week/month. As long as people are aware of the benefits of sampling at different times when possible/convenient, it is more effective overall to accommodate what works for them; that way, they are more likely to continue to collect data.
Consider volunteer’s capacity and commitment: Note that whilst more frequent data collection is generally better (e.g. weekly or fortnightly water quality testing provides a more comprehensive understanding of conditions than monthly testing), it is important that volunteers take on a level of commitment that is manageable and (in most cases) sustainable over the medium to long term (> 1 year). Regular monthly data collected over a period of years is likely to be more useful than weekly data collected for just a few months. Proactive communication with your volunteers is important to ensure that they are comfortable with their level of commitment: volunteers rarely express a growing dissatisfaction; they are more likely to just stop. Effective communication with volunteers will also help to redirect resources between sites or survey methods, e.g. if a particular site is important and warrants fortnightly data collection, then consider ‘doubling up’ by assigning two (or more) volunteers to sample the site at monthly intervals, but staggered so that the site is visited by alternate volunteers every other week.
Consider the implications of your monitoring on ecosystems: Another situation when more is not necessarily better is with biological survey methods that are intrusive or destructive, e.g. kick sampling for invertebrates is very disruptive to the habitat you are trying to assess. Populations of invertebrates that are affected by the sampling must be allowed time to fully recover before the next survey.