Key points

  • Good quality elevation data is essential for drainage design and installation
  • Digital technology products must be ground truthed in the paddock
  • Elevation, modelling, EM, NDVI and yield information collated in a layered format can provide a powerful information base for drainage planning
  • Drain locations should be GPS mapped during installation

Background

Digital and Precision Ag technologies help to streamline the process of design and installation of surface and subsurface drainage. Digital technologies also provide an understanding of what is driving crop and soil variability and thus often provides insight into variations in soil type, drainage, under or over irrigation, and differences in nutrition or disease occurrence. There are several digital technologies that can be used to understand this variability and aid drainage design.

Elevation data

The starting point for Precision Ag is guidance that provides the basis for geo-referencing. Real Time Kinetic (RTK) tractor guidance systems that are fitted to most new tractors, also have the capability to log elevation data to within less than 5 cm accuracy. Alternatively, contractors can usually map elevation to within 2 cm accuracy using a global positioning system (GPS) mounted on a quad bike ATV or side by side UTV and using an in-the-paddock base station (Figure 20). Another source of elevation data is airborne LiDAR (Light Detection and Ranging) that covers many agricultural areas of Tasmania and has 0.8 m elevation accuracy. All LiDAR held by Land Tasmania including the Sustainable Timber Tasmania (Forestry) data is freely available to download from the web by accessing the Elevation Foundation Spatial Data site (ELVIS) at: http://elevation.fsdf.org.au/. LiDAR elevation data may not be accurate enough for small increments of height or gradient. Where more accurate elevation data is required for operations such as land planning, accurate elevation data is best collected by a contractor using an on-site RTK base station (Figure 20). When collecting this data it is best to NOT have an implement on the tractor, drive at an even speed and make sure your tyre pressures are even. Any one of these factors can cause errors in the data and a lot of these smaller issues can compound into larger errors. Any spatial data collected on the farm will need clearly identified paddock and farm boundaries to assist with data collection. This can be done in Google Earth by creating a polygon around the paddock/area of interest and labelling features with field ID’s or by mapping paddock boundaries with a tractor guidance system.

Figure 20. GPS and elevation data collection set up on a quad bike (Photo by Will Wishaw) and an in the paddock base station (right) for accurate elevation data and machinery control.

Figure 20. GPS and elevation data collection set up on a quad bike (Photo by Will Wishaw) and an in the paddock base station (right) for accurate elevation data and machinery control.

Computer modelling of water flow

The detailed elevation data is loaded into digital elevation model computer software (e.g. Trimble, Ag Leader, Terra, OptiSurface, Vision Ag) to produce maps and a model of surface water flows that informs the basis of drainage design, levelling operations and surface drainage works (Figures 21 & 22). The software uses the data collected when surveying, to model outcomes of a range of events using simulation. Large rainfall events can be entered with known factors such as topography, soil type and saturation point to simulate what occurs in an event over varying time frames. The modelling indicates water flow paths, potential ponding and paddock outfalls. This software also allows variations in surface drainage depth and placement, raised bed configurations and even tractor tram lines to model what happens under different rainfall scenarios. Simulation will give options to change where field drains intersect tram lines or raised beds to alleviate ponding. The farmer, adviser and drainage contractor need to work closely together to confirm the design and associated budget. Contractors can then load the chosen drainage designs into their GPS equipped machinery to guide their operations from the cab and to accurately control the equipment. The contractor will be able to map the true location of any installed drainage pipes or ditches. This information should be retained for future maintenance or drainage expansion projects.

Following existing natural depressions and drainage lines where possible will help preserve elements of the natural drainage system and minimise risk of impact to the environment.

Figure 21. Elevation data 				Figure 22. Modelled water flow paths

Figure 21. Elevation data Figure 22. Modelled water flow paths

Crop sensing imagery

Remote crop sensing imagery can often be used to identify crop variability associated with wet areas. The normalized difference vegetation index (NDVI) measures the difference between visible and near-infrared (NIR) light reflectance from vegetation to create a snapshot of photosynthetic vigour or ‘greenness’. The sensors are responsive to both crop biomass (amount of vegetation) and crop colour (which relates to chlorophyll concentration and/or nutrient concentration). A darker green crop gives higher values than a paler green crop for the same given biomass. Darker green areas in summer indicate greater plant vigour, which may be associated with waterlogging at other times of the year (Figure 23). Freely available NDVI images and software are available that provide low resolution satellite images, however higher resolution (sub 1m resolution) images are available at a cost. High resolution multispectral imagery can also be collected using plane, drone or ground-based rovers. With better access to mobile sensors and drones in the near future, this cost will most likely decrease. Regardless of the source or quality of the images the relationship between NVDI and the cause of differences in plant vigour must be ground truthed in the paddock.

Figure 23 shows an NDVI image of an onion paddock. The image clearly shows the amount of biomass/vigour difference across the crop. Ground truthing determined that the red areas on the image were waterlogged for longer after a rainfall event than the green and yellow areas. The soils in the red areas were heavier clays with low permeability resulting in surface water ponding, waterlogging the onions and stunting their growth. This information was used for future drainage planning (Gibson 2016).

Figure 23. NDVI imagery of an onion paddock with red areas having less plant vigour than green and blue areas  due to waterlogging. Image by Greg Gibson.

Figure 23. NDVI imagery of an onion paddock with red areas having less plant vigour than green and blue areas due to waterlogging. Image by Greg Gibson.

Soil sensing imagery

Maps of soil variance can be produced by electromagnetic induction (EM) using a specialised sensor (EM38) towed behind a vehicle across paddocks (Figure 24). This can be done at the same time as collecting elevation data. EM data measures the apparent soil electrical conductivity which is associated with differences in the soil due to water content, the amount of clay, areas of rock, levels of salt or levels of other plant nutrients. Essentially, EM maps display variations in subsoil conductance without explaining why. This information is however very useful for identifying major changes in subsoil type, and identifying different soil management zones for seed varieties, crop inputs, drainage and irrigation scheduling.

Figure 24. EM map showing zones of difference due to soil/water properties. Image by Reuben Wells.

Figure 24. EM map showing zones of difference due to soil/water properties. Image by Reuben Wells.

Drones or multi rotor unmanned aerial vehicles (UAVs)

One of the advantages of drones over ground-based sensing systems is that drones are not constrained by barriers on the ground such as fences, drains or irrigators. They can fly over wet, rough ground and reach places other equipment, such as a side-by-side, cannot. Drones can ‘over-map’ a paddock boundary so that elevation data is captured on potential outfalls and drainage through neighbouring paddocks or farms. Drones help avoid biosecurity concerns because you don’t have to enter the paddock to gather the data. Drone based survey tends to be faster than vehicle-based approaches. Drones collect data in a grid pattern, so that there are no gaps in the spatial data, and the low and high spots can be determined with accuracy. Drone imagery is most useful over very flat ground with little elevation difference as the data is very accurate when the photogrammetry is collected with an in the paddock RTK base station. High-resolution imagery from a visible-colour camera on a drone can be used to generate a highly accurate 3D digital terrain model which is useful for designing a drainage system. There is a large amount of data collected by drone imagery that must be cleaned and simplified for use in the computer modelling to check on catchment flows and rainfall scenario testing that will impact drainage design. Drone imagery can be thought of as a ‘foundation data set’ as it can be utilised for several farm planning functions, not just drainage design.

Drone imagery can be used to locate old underground drainage pipelines with the use of camera or thermal imagery. The soil directly over the drain line will dry out first, be lighter in colour and reflect more light, and have a different temperature compared to the soil between the drain lines which is darker and might also be a different temperature. The soil warms up at different speeds depending on whether it is dry or wet. The soil directly on top of the drainage pipes will have a different temperature and would appear as a linear feature on the thermal infrared images. High resolution multispectral crop imagery can be captured using a drone mounted sensor to produce images of NDVI. Drones do have their limits as they don’t work well with wind and trees. If there’s too much wind, the drone won’t be able to stabilize itself at a certain elevation and trees can interrupt their flight path. Specialised drone contractors have up to date drone hardware, the appropriate computer software, the necessary camera accessories as well as experience in operating the drones to get the best input for drainage design.

Ground truthing

Digital data including elevation, Lidar, EM and NVDI maps, together with remote images from either drones or satellites provide valuable information for designing and installation of drainage works. However, none of these tools replaces the need for ground truthing to understand what any of the sensed data means in the paddock or crop. This can involve investigating soil profiles by digging holes in the different zones and assessing texture, colour and drainage, and/or by collecting soil samples from each soil management zone for laboratory testing. The digital information will not necessarily indicate where the best placement of a drain should be, particularly when strategic drains are being planned that will intercept surface or ground water flows. There is no substitute for walking over the ground when it is at its wettest to help an experienced drainage practitioner decide on the best placement of drains. The digital technology does not replace the input of farmers, agronomists or consultants, but it can help to do a better job in a shorter time and provide information that helps to design a drainage system that will reduce variability in the paddock.

Installation

Machinery grade control by GPS or laser is essential for drain installation as it maintains a positive grade by regulating the hydraulics on the boom of the drainage machinery. If the machine moves up and down while crossing undulating ground, the control unit will either lift or drop the boom to maintain the desired predetermined grade. This can only be achieved by using information provided to the unit by either GPS or laser. Laser control is widely used but GPS is becoming more popular. Laser systems use a light emitter on a field-based tripod that the receiver, mounted on the drainage machine, captures. The emitter determines the grade and depth of pipe installation, and the receiver controls the hydraulics on the drainage machine. If in undulating ground, multiple grades will need to be calculated ensuring that there is no negative grade. This reduces the risk of pipe silting or drainage failure (Gibson 2016). Water will not flow uphill!