Printer Friendly

The pugometer: an evaluation of a new tool for assessing treading damage of pasture soils and comparisons with other methods.


Treading damage of paddocks by grazing livestock degrades soil quality and increases the loss of nutrients and the transport of pathogens in surface runoff (Drewry et al. 2003; Kurz et al. 2006; Bilotta et al. 2007). Intensive pugging events can also cause considerable damage to pasture, which can result in large reductions in pasture utilisation and yield (Home and Hooper 1990; Nie et al. 2001 ; Menneer et al. 2005; Drewry et al. 2008; Phelan et al. 2013; Tunon et al. 2014). Although these effects of treading damage have been researched, accurate but practicable methods to measure or assess the magnitude and extent of treading damage have proven to be more elusive. Land managers require a quick and simple procedure for assessing the severity of treading damage, whereas researchers need a more spatially aware and quantitative measurement of the extent of pugging damage.

Cattle treading can result in hoof indentations that penetrate or rupture the soil surface (Scholefield and Hall 1985; Bilotta et al. 2007). Thus, measurements of soil roughness or disruption may be used as indicators of the severity of treading damage to soil. Four means of measuring treading damage in this manner are the depth of pug method (Nie et al. 2001; Zegwaard 2006; Tunon et al. 2014; Little et al. 2015), the pin and profile meter method (Davies and Armstrong 1986; Betteridge et al. 1999), the more frequently used roller chain method (Nie et al. 2001; Pande et al. 2002; Drewry et al. 2003; Zegwaard 2006; Tunon et al. 2014; Little et al. 2015) and the visual scoring method (Sheath and Carlson 1998; Nie et al. 2001; Zegwaard 2006; Little et al. 2015).

The depth of pug method typically involves manually measuring (with a ruler) the depth of 20 pug marks randomly selected within a given area. This method has been used on research trial plots ranging in area from 55 to 300 m2 (Nie et al. 2001; Zegwaard 2006; Tunon et al. 2014; Little et al. 2015). The ruler is placed in the deepest part of each hoof imprint and the length to the field surface, i.e. the lip of the hoof depression, is measured. The average depth of pug in the area of interest is calculated and reported.

The roller chain method was originally developed by Saleh (1993) to measure soil surface roughness caused by wind and soil erosion and tillage processes. More recently the chain technique has proved to be a useful means of measuring treading damage (Nie et al. 2001; Pande et al. 2002; Drewry et al. 2003; Zegwaard 2006; Tunon et al. 2014; Little et al. 2015). The roller chain method utilises the ability of the chain to closely follow the micro-contour or outline of the damaged, disrupted soil. As soil surface roughness increases (treading damage becomes more severe), the distance between the two ends of the chain, measured at the soil surface, decreases. Although there is no published standardised index to interpret chain measurements (Zegwaard 2006), several attempts have been made to correlate the percentage reduction in chain length to other indicators of treading damage. Zegwaard (2006) compared the percentage reduction in chain length to an arbitrary index of soil damage, which was based on roughness classes of 1-5. Zegwaard (2006) assigned chain length reductions of 0-5% to the roughness class of 1 (slightly rough), whereas chain length reductions >20% were classified as 5 (extremely rough). However, most studies tend to use the roller chain method solely as a way to quantify a trend of increasing or decreasing soil damage, rather than attempting to establish a set index or categories of treading severity (Nie et al. 2001; Tunon et al. 2014).

The visual scoring method uses treading damage criteria based on one or more of the following: a series of photos illustrating the range of treading damage, pugging depth, the percentage of bare ground, or percentage of an area damaged (Sheath and Carlson 1998; Nie et al. 2001; Zegwaard 2006; Little et al. 2015). Visual scoring methods can be performed at quadrat (Nie et al. 2001), small research plot (Zegwaard 2006; Little et al. 2015) or paddock scale (Sheath and Carlson 1998). The visual scoring method needs to be undertaken by a practiced user who is familiar with the particular criteria used to gauge treading damage. Depending on the type of criteria, the user will need to walk or stand and observe areas of interest and record the subjective measure of the treading damage observed.

The pin and profile method was first designed and used for cultivation studies by Kuipers (1957). This method uses several metal pins which are held within a metal frame and lowered onto the damaged soil. As the pins drop they 'mirror' an image of soil micro-contour that can be measured for height or graphed manually by tracing the pin heights on a white board behind the pins. This forms a visual image and a measurement of treading damage.

All the methods discussed above have been reported to be effective in measuring the degree and intensity of pugging damage. For example, studies such as Nie et al. (2001), Tunon et al. (2014) and Little et al. (2015) report that depth of pug, roller chain and visual scoring methods have been able to measure a wide range of pugging damage (P<0.01 to 0.05). The pin and profile method has been used in a grassland treading study in England by Davies and Armstrong (1986) and in a study of treading damage in New Zealand hill country by Betteridge et al. (1999). Both these studies were able to differentiate their treading treatments according to the intensity of damage using this method. However, this particular method, although accurate, has been criticised as being too impractical for routine use in the field, being labour intensive and time-consuming (Saleh 1993; Ward and Greenwood 2002; Thomsen et al. 2015). Therefore, the pin and profile method is not commonly used to measure treading damage.

The three more common methods (depth of pug, visual scoring and roller chain) are all able to characterise the severity of treading damage, but to the best of the authors' knowledge, they have not been directly correlated against each other. However, there have been studies that investigated the correlation between the pin and profile method and the roller chain method on cultivated and/or rain-damaged plots (Saleh 1993; Jester and Klik 2005; Thomsen et al. 2015). These studies have shown strong agreement between the two methods with [R.sup.2] values of 0.76-0.96

Although the methods discussed above are relatively simple and somewhat successful, they are not without their limitations. As such, no quick and reliable method has been developed that is capable of assessing the variability in the extent and severity of treading damage, in a quantitative manner, on areas from the small (plot or part-paddock) to large (paddock or whole-farm) scale. Any such method will need to be sensitive enough to discern the spatial variability in treading damage that can occur both between and within paddocks and this technique should be able to capture or record this information in an automatic and spatially explicit manner. To this end, a new spatially aware tool called the pugometer has been developed to help better quantify treading damage.

The objectives of this paper are to describe the pugometer and compare it with the three most commonly used methods for assessing pugging damage. On the basis of this comparison, a series of recommendations will be developed to guide the selection of the most appropriate technique to employ to gauge the extent of treading damage in a range of situations. The use of the pugometer will also be demonstrated in a small study of the rate of recovery of surface roughness, associated with treading.

Material and methods

Experimental procedure and site details for the comparison of methods

The experiment was conducted on three paddocks at Massey University's Dairy 4 farm near Palmerston North, Manawatu, New Zealand (NZMS 260, T24, 312867). The paddocks have flat topography (~<3% slope). The soil in the paddocks is Tokomaru silt loam, which is classified as an Argillic-fragic Perch-gley Pallie soil (Hewitt 2010). A detailed description of soil physical properties is provided by Scotter et al. (1979). The soil is naturally poorly drained and consists of a weakly to moderately developed brown, silt loam A horizon to a depth of 250 mm, a weakly developed, grey, strongly mottled, clay loam B horizon to 800 mm and a C horizon of highly compacted, pale grey, silt loam fragipan, which acts as a natural barrier to drainage (Scotter et al. 1979; Shepherd 1984). Therefore, this soil is subject to seasonal waterlogging and thus susceptible to treading damage. All paddocks are mole and pipe drained, and grow a mixed sward of perennial ryegrass (Lolium perenne) and white clover (Trifolium repens). The three paddocks were chosen because they had recently (within 2 days) been grazed under relatively wet conditions and this grazing resulted in a range of levels of treading damage.

The assessment of treading damage was conducted using four different methods: visual scoring (VSM), roller chain, depth of pug and the pugometer. Five areas, representative of each of the VSM scores, 1-5, were identified across the three paddocks. This gave a total of 25 sites. At each of the 25 sites, a 0.60 [m.sup.2] (0.6 m x 1.0 m) quadrat was placed on the ground in an area where the damage was very consistent with the visual score for the site. Treading damage was measured in each of the 25 quadrats using the depth of pug, reduction in roller chain length and pugometer methods. The shorter sides of the quadrat were marked so that it could be used to define three 1 m transects, each 0.2 m apart. The reduction in roller chain length and pugometer measurements were made along these transects. There were a total of 75 transects measured over the 25 quadrat placements. The time taken to perform 10 measurements of each method was recorded.

A comparison of methods to quantify the severity of treading damage

Visual scoring method

The VSM was developed by constructing a catalogue of treading damage on Massey University's Dairy 4 farm over the winter/spring of 2014. This follows the procedure described by Little et al. (2015) and Zegwaard (2006). A log was kept of a wide range of treading events over two winter/spring periods. This log recorded a description of damage, example photos of damage and the location of the damage on a farm map. The visual indicators of treading damage were observed by two persons along with the description of the damage. From the catalogue, five levels of damage were identified. The categories varied according to the visual depth of indentations (pug marks) in the soil, the presence of hoof smears (where the hoof of the animal had slid across the soil surface), and the degree of surface disruption. The VSM scale goes from score 1, which denotes minimal treading damage to score 5, which describes severe treading damage. The VSM developed here is categorised by the extent and severity of damage according to a five-point scoring system outlined in Table 1.

Following development, the VSM was used to assess treading damage on the farm, a plot-scale research trial, and as part of the study reported here. Treading damage to an area is characterised by comparing the state of the surface soil and pasture with the series of reference photos and the description of the VSM scores (Table 1). If the area of damaged pasture has two different levels of damage, then an intermediate score can be assigned to reflect this e.g. if the area observed has ~40% of score 2 damage and 60% of score 3 damage then the area could be scored slightly in favour of the 3 category, say as a 2.6. However, the experimental areas were chosen that represented a single score level, and thus no intermediate score was recorded in this trial.

Roller chain method and chain index

The roller chain (L1) used in this experiment was 1 m in length and made of individual links that were 22.85 mm in length. The chain measurements were performed along the three transects of the quadrat. Starting at the top left mark on the quadrat, the chain was laid along the ground with care taken to ensure that it followed the outline of the pug marks across the transect line. The horizontal distance covered (L2) was then measured; this distance decreases as soil damage increases. As the difference between L1 and L2 is related to the degree of treading damage, values obtained using the chain method are presented as a chain length percentage reduction (CLPR), which is the percentage difference between the effective length (L2) and the actual length (L1) i.e. CLPR = (L2/ L1) x 100. The chain was then repositioned at the other end of the same transect (top right-hand side) and the same process was performed. The two measures along the transect are then averaged and recorded in sequential order. This process was then repeated along the other two transects in the quadrat to provide a total of six measurements that were averaged to also give a single quadrat mean value.

Measured depth of pug

As the quadrat used in this experiment was relatively small (0.60 [m.sup.2]), the depth of pug method was modified slightly. In other studies, the depth of 20 randomly selected pug marks is measured. As there were fewer than 20 individual pug marks in each quadrat, the depths of the three deepest and three shallowest pug marks in each quadrat were measured. The depth of the pug mark was measured at the front of the pug mark using a 30-cm ruler. The depth of the six measured pugs for each quadrat was then recorded and averaged to give a mean depth of pug in the quadrat.


In order to quantify the magnitude and spatial variability of treading damage, a new tool called the pugometer was developed. The pugometer (Fig. 1 and Fig. 2) is a prototype device that has been designed based on the concept of the pin and profile meter described by Kuipers (1957). The pugometer is a mobile, handheld, global positional system (GPS)-enabled, electronic surface roughness meter, which is used to quantify treading damage. The main advantages of this apparatus are that it can quickly give a quantitative measure of the severity of treading on a given area and the results can be mapped using geographical information system (GIS) software to illustrate the spatial variation in treading. The pugometer is an alternative to more traditional measures of soil surface roughness, such as the roller chain method or the depth to pug method.

The pugometer consists of 10 stainless steel pins (0.6 cm diameter) fitted inside an aluminium frame at a spacing of 5 cm. Each of the pins has a plastic reflecting disk set on the top of it to face an infrared sensor (Model = Sharp GP2D120). The measurement is based on the intensity of the reflected light; the more emitted light the sensor detects, the nearer the disk is assumed to be. The pins slide up or down to conform to soil surface irregularities, with each pin measuring the distance between the bottom of the device and the soil surface. The further the sensor from the reflecting plate, the deeper the pugmark being measured. Soil penetration is insignificant as the pins are not spring loaded and smooth-edged, meaning they do not penetrate the soil, but rather sit on the surface of the soil.

The pins can slide up to 10 cm, i.e. they can detect pug marks to this depth. As the infrared sensor used is most accurate for depths of 0 to 10 cm, this was selected as the range of measurement depths. The dimensions and weight of the pugometer are as follows: length, 53 cm; height (from bottom of pugometer to the top of the handle), 76 cm; width, 5 cm; and a weight of 2.9 kg, thus making it small and light enough for prolonged usage.

As the sensors measure the reflected intensity of light, which is recorded as a voltage, these readings need to be calibrated to a measure of distance. Therefore, a calibration curve was created that converted light intensity to a measure of centimetres. This was developed by positioning the pugometer flat on the ground so all the pins were retracted in the device and a measurement was taken. Then 1 cm (in height) blocks where placed at either end of the pugometer so each edge of the device could rest on them and the pins would drop 1 cm to the ground surface. A reading was taken and recorded for each of the 10 pins. The height of the blocks was increased to 2 cm above the ground to allow the pins to drop 2 cm. Again a reading was taken and recorded. This process was repeated a total of 10 times so at the final measure, the pugometer was elevated 10 cm and the pins had dropped 10 cm. Each of the 10 pins was then individually calibrated to report light intensity as centimetres of pin drop.

The pugometer's design enables it to sample 10 pin depths at equally distanced positions, which are averaged to provide a single measurement. Not all pins during a measurement extend into the bottom of pug marks i.e. two pins may be in a pug mark but eight may not, therefore, the pugometer does not give a measure of individual or average pug depth, but rather it provides an indication of surface roughness.

There is a data logger connected to the pugometer by electrical wires. This data logger unit includes a recording secure digital (s.d.) card chip, Bluetooth chip and GPS circuit board. The data logger can be carried in a bag so that the only part of the device being held in the hands of the user is the pugometer itself. The apparatus is operated by first lifting the device and then positioning it on the soil. Once the device is resting on the soil, a button on the device handle is pushed to make a measurement recording. The voltage reading is recorded on the SD card in the data logger unit. The GPS location is also recorded each time the button is pressed. The device is then lifted and repositioned as many times as the user wishes. With data from multiple positions, a detailed picture of the extent of treading damage emerges.

For the measurement comparison experiment, four equally spaced readings were made with the pugometer along the same 1-m transects used for the roller chain method. The four measures were averaged for each transect. As for the roller chain method, this procedure was repeated for three transects per quadrat, which were averaged to provide a single mean value for each quadrat. This process was repeated for all 25 experimental sites.

Experimental procedure and site details for the pugometer demonstration

The experiment was performed on the same farm as experiment 1, but in a different paddock. This paddock was chosen because it had four areas of contrasting treading damage as a result of strip grazing in-calf heifers at an average stocking rate of 215 head/ha over a 4-day period (15-18 July). A photo of the paddock was taken by an un-manned aerial vehicle (UAV) after grazing (Fig. 3). The first grazing was conducted in the rain and then the soil dried over the next 3 days. The strips were grazed first at the top end of the paddock (grazing strip one) and lastly at grazing strip four. The individual transects were identified and marked, with fencing standards, at intervals of 7.5 m so that the transects could be identified for later measurements. The transects started 10 m from the paddock entrance ways and ended at the edge of a gully in the top left-hand corner of the paddock. The side of the paddock that included the gully was excluded from the experiment so that all measurements were made on the same topography.

The damaged area within each strip was visually scored using VSM on 19 July, which was 1 day after the final grazing. The VSM scores were 5, 3, 2 and 1 for grazing strips one, two, three, and four respectively (Fig. 3). Pugometer measurements were made at approximately every 4th pace along each transect, which gave ~36 measurements/ transect. This was carried out on 22 July, which was 4 days after final grazing. In order to evaluate the potential of the pugometer to characterise recovery from treading damage, surface roughness along the transects was re-measured at 49 and 104 days after the first grazing events. Once all the measurements have been recorded, the data is written to a csv file and imported into a geodatabase in ArcMap 10.5 software (ESR1). Using the Geostatistical Wizard, the point layer is used to create a simple prediction surface using the kriging interpolation method. From this, maps revealing the spatial variability in treading damage were generated.

Statistical analysis

As the VSM is a subjective measurement, linear regression could not be used to compare this method with the other three methods, thus all analysis comparing the VSM used an ANOV A, whereas the other three methods were compared using linear regression.


A comparison of methods

The approximate time taken to use each of the techniques compared here is presented in Table 2. The fastest measurement of pugging damage was the subjective VSM, as it has only one measurement and only takes the time required to observe the area and then manually record the VSM score. Each area was easily viewed and a VSM score decided upon within a period of 20 s. The pugometer was the second fastest method even though it had the greatest amount of measurements per quadrat. Each pugometer measurement took only approximately 5 s. The average time to measure a quadrat with the pugometer was 60 s. The depth of pug was the third fastest method, taking a total of 90 s to complete a quadrat, with the average time taken to identify and measure each individual pug being ~15 s. However, although the deepest pugs in a quadrat were easily identifiable, the three shallowest took slightly longer and increased the average time taken.

The slowest method was the roller chain method, which at 540 s/quadrat was very time consuming compared with the other methods. This method was slow because of the care needed to ensure that the chain was moulded to the contour of the soil surface before the reduction in chain length could be measured and then manually recorded.

The complete set of pugometer measurements (75 transect points from 25 areas) are compared with the corresponding values obtained using the roller chain method in Fig. 4. A significant positive correlation was found between the two methods ([R.sup.2] = 0.72, P < 0.01 n = 75). At lower levels of treading damage, there was very good agreement between the roller chain reductions and pugometer values. However, the differences between the two methods were generally larger at higher levels of pugging i.e. values above ~20% reduction in roller chain length and a 3.5 cm pugometer score.

There was a strong positive correlation between the average values obtained for the quadrats using the pugometer, roller chain and depth of pug methods (Fig. 5). In all three comparisons, the strongest relationship was between the depth of pug and the roller chain methods ([R.sup.2] = 0.871, P < 0.01, n = 25), and between the pugometer and the roller chain method ([R.sup.2] = 0.867, P < 0.01, n = 25). This comparison at the quadrat level suggests much better agreement between the pugometer and roller chain methods than that observed at the transect level (Fig. 4), presumably as a consequence of averaging the values across the three transects for an overall quadrat value.

The pugometer and the depth of pug method had the lowest correlation coefficient ([R.sup.2] = 0.745, P < 0.01, n = 25). There are two possible explanations for this. First, the infrared sensors used in this prototype pugometer restrict the length of the steel rods to 10 cm. Accordingly, pug marks greater than 10 cm would only be recorded as 10 cm by the pugometer. Second, the pugometer has 10 steel rods that are separated by 5 cm spacings. Therefore, the pins will not necessarily coincide with the deepest part of the pug mark, which is often the front of the pug. If the pugometer pins routinely missed the bottom of the pug mark, then its values will not be as strongly correlated with the depth of pug method, which measured the deepest part of the pug mark.

There was a significant positive relationship (P < 0.0001) between the VSM and the roller chain, pugometer and depth of pug methods (Fig. 6). All three methods were able to clearly identify a difference (ANOVA, P<0.05) between the levels of treading damage as assessed by the VSM scores with at least one score difference between them, i.e. 1 vs 3 or 2 vs 4 etc. However, it was more difficult to significantly discriminate between consecutive VSM scores for all three methods. Values for the roller chain method were the closest match to VSM scores, and it was able to significantly (P<0.05) distinguish between all VSM scores except between a score of 3 and 4.

Use of the pugometer to assess treading damage

The GIS interpolated data as recorded by the pugometer provide a graphic illustration of the spatial variation in treading damage (Fig. 7). The area grazed during the wettest soil conditions is clearly identified as having a higher degree and extent of treading damage (top section; Fig. 7). Damage associated with cow and vehicle traffic through the gateway is also clearly identified in the bottom right-hand side of Fig. 7. The pugometer also suggests that three of the strips sustained relatively little treading damage apart from some poorly drained areas in the central part of the right-hand side of the paddock. This points to the relatively quick drying of the surface soil so long as the mole-pipe drainage system is fully operational.

Treading damage across the four grazing strips was also assessed using the VSM. A series of GPS referenced VSM scores, and accompanying photos, are shown in Fig. 7. Photo (a) shows a treading damage score of 4, which was also identified by the pugometer as an area with a high level of soil damage. Photo (b) shows a score of 0.5 indicating that there were minimal imprints on the soil surface, which again agrees well with the pugometer assessment. Photos (c) and (d) show VSM scores of 3 and 2 respectively, and the pugometer's evaluation of these intermediate levels of treading damage. It is noteworthy that there is no easy way to illustrate the variability in treading damage across the area using the VSM scores.

Although aerial photography (Fig. 3) could be used to identify and quantify damage at the paddock and farm scale, in this study it was only able to capture some of the more extreme treading damage (i.e. VSM scores of >4). More detailed methods of photo analysis were outside the scope of this paper and were not investigated, but could be employed to characterise treading damage in the future.

It has been observed that the surface roughness, associated with treading damage, can recover quickly (Sheath and Carlson 1998), with reports of recovery times between 87 to 165 days, depending on the characteristics of the soil (Elliott et al. 2002). Factors that contribute to recovery include wetting and drying processes, rainfall induced subsidence of soil, earthworm activity and cattle hooves scuffing and knocking raised lumps of soil to fill hollows (Singleton and Addison 1999). This experiment evaluated the use of the pugometer to monitor the recovery of treading damage, as measured by reductions in soil surface roughness. Fig. 8 shows the pugometer assessments of recovery from treading damage in the strips at intervals of 49 and 104 days after the initial wet soil grazing. It would appear that even the relatively severe surface roughness had disappeared after a period of 104 days. It is important not to read too much into this value of 104 days as this recovery period was unique to this particular treading event. The main point here is that the pugometer allowed this information to be collected rapidly, i.e. in less than half an hour per measurement period and, therefore, will be a useful tool for more comprehensive studies of the recovery of surface roughness following treading damage. None of the methods previously discussed are able to provide and automatically record this level of spatial detail within a comparable timeframe.


Comparison of the four methods

Treading damage, as inflicted on soil by grazing animals during wet periods, is difficult to quantify on a large scale, particularly in a quantitative way for research purposes. In the first instance, surface roughness is most commonly used as an indicator of the extent of treading damage. Although multiple methods to quantify surface soil roughness have been compared on tilled arable land (Thomsen et al. 2015), the current study is the first known comparison conducted on grazed dairy pastures.

All four methods were able to quantify treading over the range of damage levels observed during this study. Despite the marked differences between the methods, the values that they generated were strongly correlated, i.e. there was strong agreement between methods in their assessment of treading damage. Perhaps this is unsurprising given that three of the methods are commonly used to measure treading damage. However, the newly developed pugometer produced similar results to the other methods tested in this paper, which is a useful new finding.

At a detailed level, the pugometer, roller chain, and depth of pug methods all compared well against the subjective VSM. However, these other methods were not able to differentiate between consecutive VSM scores (i.e. one score unit). Although the VSM is subjective, it is a rapid, practical and reliable method for assessing soil damage up to a paddock scale. Moreover, in the human eye, it employs one of the most powerful instruments known. The scorecard presented in Table 1 provides guidance to help provide consistency and reduce variation between different users.

That assessment of treading damage can be laborious and time consuming is a common perception amongst both researchers and farmers. Therefore, the time taken to complete an assessment of treading damage is an important feature of any particular method. One of the biggest differences between the methods was the time taken to conduct measurements and record them, which highlights the advantage of rapid techniques such as the pugometer and VSM. The depth of pug method is slightly quicker than the roller chain, but neither method is practical for measuring treading at a paddock scale. Like the VSM, the pugometer provides a rapid method for assessing treading up to the paddock scale, but the latter has the advantage of being able to automatically record spatial variation in soil damage across an area. This is useful because treading damage can be highly variable, even within the same paddock, as seen in Fig. 7. This variation arises for a range of reasons including paddock breaks being grazed on different days with varying soil moisture conditions, variation in drainage across a paddock, different soil types, and differences in cow traffic and animal behaviour.

To illustrate the differences in the time and number of measurements it would take to conduct the four different methods over a larger area than a quadrat, the following example is used. Take a hypothetical square 100m x 100 m area (1 ha), which has been badly damaged by grazing cows. Table 2 suggests that the VSM is the fastest method. If a walking pace of 3km [h.sup.-1] is used to walk an 'M' shaped transect, a distance of ~500m would be covered in 10 min. The total time to complete the VSM would be 10 min. To cover the same transect using the pugometer, a reading would be taken at every 4th pace (~4m): this would give 125 readings and take 20 min to complete (10 min to walk the area plus 10 min to complete the readings). For this larger area, the depth of pug and roller chain methods could be performed at random or representative points on this same transect. If there is the same 20-min period allowed (like the pugometer) for the depth of pug method, and it takes 10 min to walk the transect then there would be time to measure the depth of pug marks in seven quadrats. As it takes 90 s to complete a roller chain measurement, this method would only be able to take seven measurements in 10 min. This reduction in sample size limits the ability of these two slower methods to accurately quantify spatial variability. Whether the reduction in sample size would affect the correlation of the depth of pug and roller chain methods with the pugometer and VSM or give reliable estimates of treading damage over a larger area was not investigated in this study.

One of the advantages of the pugometer is that the large number of measurements and the GPS functionality allows the user to characterise the spatial variability in treading damage. Although the VSM is more limited in its ability to record spatial detail, it can still be a practical tool for farmers to record soil damage.

Given that all methods are reasonably accurate and in agreement, the choice of method for assessment of treading damage will depend on context and area. For small research plots, where spatial variability is low and the number of measurements required is relatively small, all four techniques could be considered as acceptable methods. In large plots and paddock research areas, larger numbers of measurements may be required to capture spatial variability. In these situations, both the VSM and the pugometer would be feasible options. At a farm management level, where farmers often want to quantify pugging damage quickly (to make a decision such as whether to continue grazing paddocks in wet conditions or stand cows off), the VSM would be most appropriate given that the farmer could undertake a simple paddock walk (similar to a pasture walk) to easily and quickly determine the degree of damage on an area without the use of any apparatus. For paddock or farm scale research, where a good record of the spatial variation in pugging damage is important, the use of the pugometer would be advantageous.

Demonstration of the use of the pugometer at scale

As noted above, an important advantage of the pugometer is that it can describe treading damage over a large area relatively quickly and the GPS technology allows the automatic capture of spatial variability. These features mean that the pugometer can be used to study some aspects of treading damage in detail and answer questions such as those related to recovery time.

This work has attempted to provide guidelines to help in the selection of the most appropriate method to assess treading damage. It recognises that the requirement of farm managers differs from researchers and, to this end, has introduced a new tool, the pugometer, for use in paddock-scale research. The methods compared here all equate the seventy of treading damage to the degree of surface roughness. However, there is much more to treading damage, including its impact on soil properties and processes and pasture utilisation, regrowth, and species composition. In the future, it will be important to attempt to relate the severity of treading damage as assessed by the methods used in the present study to soil and pasture characteristics. An obvious example would be to develop relationships between VSM and pugometer scores with likely pasture growth rates following a treading event. As mentioned above, the pugometer also lends itself to studies of the recovery rate of surface roughness following treading damage.


This study presents three established methods of measuring treading damage and introduces and demonstrates a new tool called the pugometer. This study is, to the best of the authors' knowledge, the first that directly compares the common methods of measuring treading damage on dairy pastures. All methods were strongly correlated and were able to identify varying levels of treading damage. However, which method is the best choice for assessment of treading damage depends on the circumstances. For small research plots, where spatial variability is low, and the frequency of measurements is relatively small, all four techniques could be considered acceptable. For large plots and paddock research the visual scoring method and pugometer would be most appropriate. However, the advantage of the pugometer in this case is that it can capture spatial variability rapidly and automatically. This was demonstrated by its practical ability to track the recovery of surface roughness following a wet-soil-treading event.

Conflicts of interest

The authors declare no conflicts of interest.


This study was conducted through the Pastoral 21 Environment Program funded by the Ministry of Business Innovation and Employment, DairyNZ, Fonterra and Beef + Lamb New Zealand. The authors thank Isabel Tait and Quang Mai for assistance with fieldwork.


Betteridge K, Mackay AD, Shepherd TG, Barker DJ, Budding PJ, Devantier BP, Costall DA (1999) Effect of cattle and sheep treading on surface configuration of a sedimentary hill soil. Soil Research 37, 743-760.

Bilotta GS, Brazier RE, Haygarth PM (2007) The impacts of grazing animals on the quality of soils, vegetation, and surface waters in intensively managed grasslands. Advances in Agronomy 94. 237-280. doi: 10.1016/S0065-2113(06)94006-1

Davies PA, Armstrong A (1986) Field measurements of grassland poaching. The Journal of Agricultural Science 106, 67-73. doi:10.1017/S002185960006175X

Drewry JJ, Singleton PL, Boyes M, Judge A, Wheeler D (2003). Short term recovery of soil physical properties after winter grazing in the Waikato: Implications for environmental monitoring. Tools for nutrient and pollutant management: Applications to agriculture and environmental quality Occasional report No. 17, pp. 194-204. (Fertiliser and Lime Research Centre Massey University Palmerston North).

Drewry JJ, Cameron KC, Buchan GD (2008) Pasture yield and soil physical property responses to soil compaction from treading and grazing--a review. Soil Research 46, 237-256. doi:10.1071/SR07125

Elliott AH, Tian YQ, Rutherford JC. Carlson WT (2002) Effect of cattle treading on interrill erosion from hill pasture: modelling concepts and analysis of rainfall simulator data. Soil Research 40, 963-976. doi: 10.1071/SR01057

Hewitt AE (2010) "New Zealand soil classification', 3rd edn (Manaaki Whenua Press: Lincoln. NZ)

Horne DJ, Hooper M (1990) Some aspects of winter management of "wet soils". Massey University Dairy Fanning Annual, pp. 90-94.

Jester W, Klik A (2005) Soil surface roughness measurement methods, applicability, and surface representation. Catena 64. 174-192. doi:10.1016/j.catena.2005.08.005

Kuipers H (1957) A relief meter for soil cultivation studies. Netherlands Journal of Agricultural Science 5, 255-262.

Kurz I. O'Reilly CD. Tunney H (2006) Impact of cattle on soil physical properties and nutrient concentrations in overland flow from pasture in Ireland. Agriculture, Ecosystems & Environment 113, 378-390. doi: 10.1016/j.agee.2005.10.004

Little CL. Hickson RE, Martin RE, Beausoleil NJ, Cockrem JF, Kenyon PR, Horne DJ, Morris ST (2015) Beef cattle wintering systems: effects on cattle and pasture. Proceedings of the New Zealand Society of Animal Production 75, 167-171.

Menneer JC, Ledgard S, McLay C, Silvester W (2005) The effects of treading by dairy cows during wet soil conditions on white clover productivity, growth and morphology in a white clover-perennial ryegrass pasture. Grass and Forage Science 60. 46-58. doi: 10.1111/j.1365-2494.2005.00450.X

Nie ZN, Ward GN, Michael AT (2001) Impact of pugging by dairy cows on pastures and indicators of pugging damage to pasture soil in southwestern Victoria. Australian Journal of Agricultural Research 52, 37-43. doi :10.1071/AR00063

Pande T, Valentine I, Betterridge K, Home DJ. Mackay AD. Hodgson J (2002) Effect of canopy height and cattle treading on herbage regrowth parameters in dairy pastures. Dairy Farm Soil Management Occasional report No. 15, 89-93. (Fertiliser and Lime Research Centre Massey University Palmerston North).

Phelan P, Keogh B, Casey IA, Necpalova M, Humphreys J (2013) The effects of treading by dairy cows on soil properties and herbage production for three white clover-based grazing systems on a clay loam soil. Grass and Forage Science 68, 548-563. doi: 10.1111/gfs.12014

Saleh A (1993) Soil roughness measurement: chain method. Journal of Soil and Water Conservation 48, 527-529.

Scholefield D, Hall DM (1985) A method to measure the susceptibility of pasture soils to poaching by cattle. Soil Use and Management 1. 134-138. doi: 10.1111/j.1475-2743.1985.tb00976.x

Scotter DR, Clothier B, Corker RB (1979) Soil water in a Fragiaqualf. Australian Journal of Soil Research 17, 443-453. doi: 10.1071/SR9790443

Sheath GW, Carlson WT (1998) Impact of cattle treading on hill land: soil damage patterns and pasture status. New Zealand Journal of Agricultural Research 41, 271-278. doi: 10.1080/00288233.1998. 9513311

Shepherd TG (1984). Distribution and description of Yellow-Grey Earths in Manawatu and Whanganui regions. In 'Soil groups of New Zealand. Part 7, Yellow-Grey Earths'. (Ed. JG Bruce) pp. 123 (New Zealand Society of Soil Science: Lower Hutt, New Zealand)

Singleton PL, Addison B (1999) Effects of cattle treading on physical properties of three soils used for dairy farming in the Waikato, North Island, New Zealand. Soil Research 37, 891-902. doi: 10.1071/SR98101 Thomsen LM, Baartman JEM. Bameveld RJ. Starkloff T, Stolte J (2015) Soil surface roughness: comparing old and new measuring methods and application in a soil erosion model. Soil (Gottingen) 1. 399-410. doi: 10.5194/soil-1-399-2015

Tunon G, O'Donovan M, Lopez Villalobos N. Hennessy D, Kemp P, Kennedy E (2014) Spring and autumn animal treading effects on pre-grazing herbage mass and tiller density on two contrasting pasture types in Ireland. Grass and Forage Science 69, 502-513. doi: 10.1111/gfs.12055

Ward G, Greenwood KL (2002) Research and experiences in treading and wet soil management in Victoria. Dairy Farm Soil Management Occasional report No. 15, pp. 47-59. (Fertiliser and Lime Research Centre Massey University Palmerston North).

Zegwaard KE (2006) Effects of severe cattle treading on soil physical properties and pasture productivity. PhD thesis, University of Waikato, Hamilton.

J. A. Howes(iD) (A,B), J. A. Hanly (A), D. J. Horne (A), M. J. Hed!ey (A), and M. Irwin (A)

(A) School of Agriculture and Environment, Environmental Science, Massey University, Palmerston North, New Zealand.

(B) Corresponding author. Email:

Receive 17 January 2018, accepted 1 May 2018, published online 31 July 2018

Caption: Fig. 1. The pugometer, battery and data logger.

Caption: Fig. 2. The pugometer with upper pin cover folded up, exposing the infrared sensors and lightreflecting disks.

Caption: Fig. 3. The areas of the paddock that were allocated to the four grazing strips (numbered 1-4). The area of the seven measurement transects used is also shown in the orange box. The entrance ways are at the top and bottom right-hand corners of the paddock. One end of the badly damaged gully is seen in the left side of grazing strip 1.

Caption: Fig. 4. Linear regression comparing 75 individual measurements made by the pugometer and roller chain method.

Caption: Fig. 5. Linear regression comparing (a) pugometer vs the roller chain % reduction, (b) pugometer vs depth of pug and (c) depth of pug and roller chain % reduction.

Caption: Fig. 6. Analysis of variance results comparing the visual scoring method (VSM) scores (1-5) against (a) pugometer, (b) roller chain % reduction and (c) depth of pug. Values with the same letter are not significantly different (P<0.05), with error bars showing the standard error of the mean.

Caption: Fig. 7. Interpolated data from pugometer measurements. A comparison of the assessment of treading damage as made by the pugometer and visual scoring method (VSM). Photos showing actual damage with a VSM score of (a) 4, (b) 0.5, (c) 3 and (d) 2.

Caption: Fig. 8. Treading damage as assessed by the pugometer after the initial grazing damage occurred (left), 49 days after initial wet grazing (middle) and 104 days after initial wet grazing (right).
Table 1. Visual scoring method visual damage score, increasing
from levels 1 to 5, with score criteria and a close-up photo of each
score level

Damage    Score criterion                 Close up
          Minimal indentations of soil:
1         no hoof smears or surface

2         Slight indentations of soil:
          some hoof smears: very
          minimal surface disruption

3         Medium indentations of soil:
          some hoof smears medium
          surface disruption

4         Deep indentations of soil:
          deep hoof smears: medium
          surface disruption

5         Very deep indentations of
          soil: deep hoof smears:
          intense surface disruption

Table 2. Approximate time to complete measurements in a quadrat
including recording the result

Method                  No. of           Time taken      Total time to
                     measurements            per          complete a
                  taken in a quadrat   measurement (s)    quadrat (s)

Pugometer                 12                  5               60
Depth of pug              6                  15               90
Roller chain              6                  90               540
Visual scoring            1                  20               20
COPYRIGHT 2018 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Howes, J.A.; Hanly, J.A.; Horne, D.J.; Hedley, M.J.; Irwin, M.
Publication:Soil Research
Article Type:Report
Date:Sep 1, 2018
Previous Article:A lab-made method for extracting DNA from soils.
Next Article:In-vitro evaluation of rice and wheat straw biochars' effect on pyrazosulfuron-ethyl degradation and microbial activity in rice-planted soil.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters