Faeces count as an index of fox abundance

1. What they did

Red foxes are hard, and in many places impossible, to count accurately. Researchers have used den counts, hunting and questionnaire statistics, scent stations, spotlighting, track counts on snow and sand, and scat counts along transects. Only thorough den counts give real densities, and they are expensive and slow. The other methods give only relative numbers, and each has its own problems. Scat counts may be affected by diet (fruit is poorly digested) and by rainfall, but the effect of rain had never been studied.

The author set out to test how consistent scat counts are as an index of fox numbers, and how fruit in the diet and rainfall affect them:

• Transects. One transect in each of nine areas of Pisa province, central Italy, 6.2 to 14.4 km long. To avoid differences in how easily scats could be seen, all transects followed paths of similar width.
• Searching. Each transect was walked on foot once a month, from April 1992 to March 1993, and searched for fox scats: 108 transects and 982 km in all.
• Identifying and mapping. Fox scats were identified by their diameter, shape and smell, following a field guide to animal tracks and signs (Bang and Dahlström 1974). Each scat was plotted on a map, and the distance to the next scat was measured (distances under 100 m were estimated in the field).
• Diet and rain. The share of fruit and other plant matter in each scat was measured in the laboratory, and rainfall in the 20 days before each walk was taken from the nearest weather station (under 8 km away).

The index was the number of scats per km of transect.

2. What they found

Scat counts

• 1293 fox scats were collected on the 108 transects. Counts varied a lot between seasons and between areas.
• The index was highest (over 1.5 scats per km) in spring (March to May) and in July and August, and lowest (under 1 per km) from October to December. This pattern was the same in all areas.
• Yearly averages ranged from 0.34 scats per km (Collebrunacchi) to 2.97 (Monti Pisani). Monti Pisani and Berignone had over 2 per km; Vecchiano and Rio-Arbiaia were intermediate; the other five areas had under 1. These differences between areas were significant and held throughout the year.
• The two areas with the highest index were more wooded and free of hunting. The area with the lowest index had been heavily hunted in spring 1992 (hunting pressure of more than 0.5 foxes per km²), but one intermediate area (Vecchiano) also had fairly high hunting pressure (more than 0.22 foxes per km²).

How scats were spread along the paths

Scats were spread roughly at random along the transects, not clumped. The coefficient of variation of the distances between scats ranged from 0.71 to 1.89 (average 1.33) and did not differ between areas.

Rain

• Rainfall in the 20 days before a walk had a significant effect on the number of scats found: the more rain, the fewer scats.
• Rainfall varied greatly between months, peaking at over 200 mm in October, but not between areas.

Diet

• Plant matter made up 0 to 100% of the diet in different months, averaging 34.8% overall, and about 90% of it was fruit. Area averages ranged from 20.5% (Coltano) to 56.1% (Monti Pisani).
• The index was linked to the share of plants in the diet, but once the differences between areas were taken into account, diet had no significant effect. The index was therefore independent of how much fruit the foxes ate.

Why rain matters

The author considered three explanations: foxes changing where they go with the seasons; tall summer grass hiding scats; and rain breaking scats down. The first two did not fit: the pattern was the same in very different habitats, transects crossed all the main habitats, and the index was lowest in winter, when the grass was short. Rain breaking down the scats was the most likely cause. In casual observations of 19 scats of known age, scats lasted in the field with very little sign of decay until the first rain; heavy rain washed them away, and lighter rain broke them up.

3. What they concluded

• Scat counts gave consistent results through the year. The method is cheap, can be used routinely even by fairly inexperienced observers such as game wardens and students, works in heavily wooded areas where spotlighting and den counts do not, and needs no snow or sand.
• The true density of foxes in the nine areas was unknown, but the index fitted the hunting pattern: foxes seemed most abundant where they were not hunted and least abundant where they were hunted most. The author suggests scat counts could be used more widely, for example instead of or along with hunting statistics, to estimate relative fox abundance.
• Because foxes do not use latrines and defecate at similar rates at the edges and in the middle of their home ranges, their scats are spread fairly evenly, so shorter transects are enough for a stable index.
• The index cannot detect seasonal changes in fox numbers. It did not rise after the spring births (around the end of April) or fall during the main hunting season (January to April).
• Rain is the main limitation. The author suggests using the index when less than 50 mm of rain has fallen in the 20 days before a walk, and comparing only areas and periods with similar rainfall. Without knowing the rainfall, seasonal changes in the index could be wrongly read as changes in fox numbers (for example, young joining the population), so the index is not recommended for comparing seasons.
• Diet did not matter in this study area, so routine surveys could skip diet analysis, saving time and skill, except perhaps where foxes eat more fruit or where fruit eating differs between areas.
• Estimating absolute numbers from scats is confused by at least two things: foxes may prefer paths, so scats along paths may not represent the whole area; and defecation rates differ with diet, digestibility and between individuals.
• The method may also work for other species if they live in dry areas or are surveyed in the dry season, tend to defecate along paths and roads, and have scats that are easily told apart from those of other carnivores in the area.

4. How tracking was or was not used

The study is a sign survey: observers walked transects on paths and searched for fox scats. Scats were identified as fox by their diameter, shape and smell, using a published field guide to animal tracks and signs. No trailing and no other tracks or signs were used. The author also observed how long scats of known age lasted in the field and how rain broke them down.

5. Tracker or observer certification

Not reported. The paper does not say who searched the transects or identified the scats, or whether their identifications were checked. T. Volpi and S. Santini are thanked for their great help with field work and laboratory analysis. The author states that fairly inexperienced observers could use the method, but this was not tested.

6. Other technologies, skills or innovations

• Scat locations plotted on 1:5,000 or 1:25,000 maps, with the distance between successive scats used to test whether scats were clumped or random.
• Laboratory diet analysis of scats (estimated volume of food eaten).
• Rainfall records from local weather stations used to explain changes in scat counts.

7. Implications for conservation (from the paper)

• Scat counts offer a cheap, simple way to monitor the relative abundance of foxes across areas, and could be used instead of or along with hunting statistics, which depend on hunters and on attitudes toward foxes.
• Comparisons should only include areas and periods with similar rainfall, and seasonal changes in fox numbers cannot be detected with this index.