Collected data
From the 101 collars deployed, 50 (49.5%) were retrieved and 46 (45.5%) had recorded data for 10 or more days (Table 1). Loss of devices was observed in all countries, most prominently in Uganda (26 (63%) of the 41 collared devices). After data cleaning, the number of complete days used for UD calculation ranged from 19 to 85 (median 44).
The number of GPS fixes per dog after data cleaning ranged between 1,756 and 6,117 (median 4,376) in Chad, between 1,446 and 4,026 (2,897) in Guatemala, between 1,193 and 3,027 (2,809) in Indonesia and between 2,837 and 7,979 (4,477) in Uganda. A total of 1486 (3.2%) fixes in Chad, 5049 (13.2%) in Guatemala, 3495 (12.4%) in Indonesia and 1932 (2.9%) in Uganda were removed during the cleaning process.
All 46 dogs enrolled were allowed to roam freely. Fifteen were female (one from Chad, three from Guatemala, five from Indonesia and six from Uganda), 30 were male (nine from Chad, nine from Guatemala, five from Indonesia, seven from Uganda). Information on sex was not recorded for one of the animals from Uganda. Age varied between four and 48 months (median 12) in Indonesia and between three and 248 months (48) in Uganda. No reliable information for age was available for the dogs from Chad and Guatemala. All dogs, except one for which the role played was not recorded, were kept as guardians. Furthermore, 19 (41.3%) were kept as pets or companion, six (13.0%) as shepherds, three (6.5%) as hunters and two (4.3%) as source of meat.
Table 1: Summary of the GPS units collared, retrieved and usable for each study site. NA = not applicable
Study site
|
Deployed (%)
|
Retrieved (%)
|
Useable data (%)
|
Median (range) number of analyzed days
|
Chad
|
NDakonon
|
11 (10.9)
|
6 (5.9)
|
5 (5.0)
|
46 (46-62)
|
Sinetaye
|
9 (8.9)
|
5 (5.0)
|
5 (5.0)
|
56 (19-65)
|
Guatemala
|
La Romana
|
4 (4.0)
|
4 (4.0)
|
3 (3.0)
|
44 (44-47)
|
Sabaneta
|
6 (5.9)
|
5 (5.0)
|
5 (5.0)
|
36 (20-36)
|
Poptún
|
10 (10.0)
|
6 (5.9)
|
4 (4.0)
|
27 (22-34)
|
Indonesia
|
Pogon
|
4 (4.0)
|
3 (3.0)
|
3 (3.0)
|
26 (23-27)
|
Hepang
|
4 (4.0)
|
0 (0.0)
|
0 (0.0)
|
NA
|
Habi
|
10 (10.0)
|
7 (5.9)
|
7 (6.9)
|
27.5 (27-28)
|
Uganda
|
Kamuda
|
6 (5.9)
|
0 (0.0)
|
0 (0.0)
|
NA
|
Soroti
|
35 (34.7)
|
15 (14.9)
|
14 (13.9)
|
48 (38-85)
|
Total
|
101 (100.0)
|
50 (49.5)
|
46 (45.5)
|
|
HR outcome metrics
Home range size
Estimated HR sizes varied along days of observation and between dogs and isopleths (Fig. 1). Generally, for short observation periods, HR sizes changed considerable for each day of additional data considered. The longer the observation period, the higher the number of dogs that seemed to reached a plateau, with little change for each additional day of observation. For the full observation period, which varied from dog to dog, the core HR size ranged between 0.2 and 1.2 hectares (ha) (median 0.4) in Chad (10 dogs), 0.2 and 0.9 ha (0.4 ha) in Guatemala (12 dogs), 0.3 and 2.9 ha (0.4 ha) in Indonesia (10 dogs) and 0.3 and 1.5 ha (0.4 ha) in Uganda (14 dogs). The extended HR size ranged between 2.0 and 1,110.7 ha (50.6 ha) in Chad, 2.0 and 92.7 ha (19.5 ha) in Guatemala, 2.5 and 80.7 ha (5.2 ha) in Indonesia and between 1.7 and 42.9 ha (8.0 ha) in Uganda. The graphs for the 60%, 70%, 80% and 90% isopleths are presented in Supplementary Fig. S3.
Shape index
Values of shape index, which were influenced by the spatial dispersion of GPS fixes, also varied with increasing days under observation and between dogs and isopleths (Supplementary Fig. S4). Similarly, to the HR size, the plot lines for each individual dog tended to reach a plateau with increasing observation periods. At the end of the recorded observation period, values of the shape index for the core HR ranged between 0.0 and 1.0 (median 1.0) in Chad, between 0.2 and 1.0 (1.0) in Guatemala, between 0.0 and 1.0 (1.0) in Indonesia and between 0.2 and 1.0 (1.0) in Uganda. The values of the shape index for the extended HR varied between 0.2 and 1.0 (median 0.2) in Chad, between 0.0 and 1.0 (0.2) in Guatemala, between 0.0 and 1.0 (1.0) in Indonesia and between 0.3 and 1.0 (1.0) in Uganda.
Minimum required observation period
Home range size
The percentage of change in HR size for each additional day of observation was calculated for different isopleths (Fig. 2, Supplementary Fig. S4). Generally, the longer the observation period, the smaller the change for each additional day of observation, with the percentage of change tending towards zero (Fig. 2, Fig. 3). The minimum number of days after which percentages of change in HR size for each additional day of observation were of a maximum of 10% and 20% was generally smaller for the core HR than for the extended HR (Fig. 2, Table 2). Some dogs needed to be observed only for one day to reach a constant maximum of 10 % of change constantly for the core HR, and two days for the extended HR (Table 2). However, other dogs needed much larger observation periods, reaching up to 33 and 13 days for the 10% and 20% of change, respectively, for the core HR; and 53 and 34 days, respectively, for the extended HR. For 75% of the dogs, 7 and 2 days were enough to reach a daily maximum of 10% and 20% change for the core HR, whereas 26 and 13 days were required for the extended HR (Fig. 3, Table 2). For the core HR, one dog (2.3%) just reached a constant maximum value of 10% change at the end of the observation period. For the extended HR, one dog (2.3%) never reached a constant maximum change of 10% and four dogs (9.1%) reached it in the last day; one of these four also just reached a constant maximum change under 20% at the last day. The number of days needed to reach a constant percentage of change equal or below 10% and 20% for the other isopleths are presented in Supplementary Table S2.
Home range shape
Similarly to what was observed for the HR size, the longer the observation period, more stable the HR shape index was for most of the dogs (Fig. 2, Fig. 3). Therefore, with increasing observation periods, the percentage of change tended towards zero. The minimum number of days after which the percentages of change were constantly up to a maximum of 10% and 20% was smaller for the core HR than for the extended HR (Table 2, Supplementary Fig. S5). There were dogs which only needed to be observed for 1 day for a maximum of 10% of change, for both the core and extended HR (Table 2). However, some dogs needed much larger observation periods, reaching up to 53 days for the core HR and up to 63 for the extended HR, even for the 20% change threshold (Fig. 3, Table 2). For the extended HR, three (6.5%) dogs never reached a constant maximum value of 10% change, and two (4.3%) of those never reached a constant maximum value of 20% change.
Table 2: Number of observed days required to reach a constant value of percentage of change equal or under 10% and 20% for the HR size and HR shape index for the core and extended HR. Numbers presented indicate the minimum (Min) and maximum (Max), median, percentile 25 (Per. 25), percentile 75 (Per 75) and mean of the number of days amongst all dogs, as well as the number of dogs that did not reach a constant value below the chosen threshold.
HR
|
Change (%)
|
Number of days
|
Number of dogs without limits reached
|
Min
|
Max
|
Median
|
Per. 25
|
Per. 75
|
Mean
|
HR size
|
Core
|
10
|
1.0
|
33.0
|
3.0
|
2.0
|
7.0
|
5.7
|
0.0
|
20
|
1.0
|
13.0
|
1.0
|
1.0
|
2.0
|
2.5
|
0.0
|
Extended
|
10
|
2.0
|
53.0
|
17.0
|
10.0
|
26.0
|
18.9
|
1.0
|
20
|
1.0
|
34.0
|
7.0
|
4.0
|
12.75
|
9.7
|
0.0
|
HR shape index
|
Core
|
10
|
1.0
|
53.0
|
1.0
|
1.0
|
7.25
|
8.5
|
0.0
|
20
|
1.0
|
53.0
|
1.0
|
1.0
|
7.25
|
8.5
|
0.0
|
Extended
|
10
|
1.0
|
63.0
|
6.0
|
1.0
|
20.5
|
13.7
|
3.0
|
20
|
1.0
|
63.0
|
2.5
|
1.0
|
13.5
|
9.9
|
2.0
|
Comparisons of the required number of days between countries and sex
Home range size
For differences in the minimum number of days for the HR size between countries, the analysis included 46 dogs (10 from Chad, 12 from Guatemala, 10 from Indonesia and 14 from Uganda) for the 10% and 20% change of the core HR and 20% change in the extended HR; and 45 dogs (10 from Chad, 11 from Guatemala, 10 from Indonesia and 14 from Uganda) for the 10% change of the extended HR (one dog did not reach the defined threshold). No significant differences were found for the core HR size (p=0.67) and 10% change. However, significant differences were found for the core HR and 20% change (p=0.02) and for the extended HR size (10% change, p=0.03; 20% change, p=0.05).
For the analysis of the influence of the dogs' sex, 15 female dogs and 30 male dogs were included for the core HR and 15 female dogs and 29 male dogs for the extended HR. One dog from Uganda did not have data on sex and therefore was excluded from this analysis. One male dog never reached a number of days after which the percentage of change was constantly equal or under 10% in the extended HR and therefore were excluded from the analysis. The effect of sex was not statistically significant in the core HR (10% change, p=0.76; 20% change p=0.61), while in the extended HR was significant for the 10% change (p=0.012), but not for the 20% change (p=0.74).
Minimum, maximum, median and mean number of days for each country and UD isopleths are presented in Table 3, Supplementary Fig. S6, Supplementary Fig. S7 and Supplementary Table S2.
Table 3: Number of observation days required in each country to reach a constant value of percentage of change equal or below 10% and 20% in HR size for the core and extended HR. Numbers presented here indicate the minimum (Min) and maximum (Max), median, mean and percentiles 25 and 75 of the minimum number of days required amongst all dogs as well as the number of dogs that did not reach a constant value below the chosen threshold.
HR
|
Change (%)
|
Number of days
|
Number of dogs with no limits reached
|
Min
|
Max
|
Median
|
Percentile 25
|
Percentile 75
|
Mean
|
Chad
|
Core
|
10.0
|
1.0
|
33.0
|
5.0
|
2.25
|
17.75
|
11.2
|
0.0
|
20.0
|
1.0
|
13.0
|
3.0
|
2.0
|
8.75
|
5.1
|
0.0
|
Extended
|
10.0
|
15.0
|
53.0
|
24.5
|
17.25
|
29.75
|
25.8
|
0.0
|
20.0
|
2.0
|
33.0
|
13.0
|
10.25
|
21.25
|
14.8
|
0.0
|
Guatemala
|
Core
|
10.0
|
1.0
|
13.0
|
3.5
|
1.75
|
4.75
|
4.1
|
0.0
|
20.0
|
1.0
|
7.0
|
1.0
|
1.0
|
1.25
|
1.8
|
0.0
|
Extended
|
10.0
|
6.0
|
34.0
|
20.0
|
14.0
|
26.0
|
19.9
|
1.0
|
20.0
|
2.0
|
34.0
|
7.0
|
6.0
|
13.5
|
10.9
|
0.0
|
Indonesia
|
Core
|
10.0
|
1.0
|
11.0
|
2.5
|
1.25
|
8.0
|
4.4
|
0.0
|
20.0
|
1.0
|
11.0
|
1.0
|
1
|
2.5
|
2.4
|
0.0
|
Extended
|
10.0
|
2.0
|
26.0
|
11.0
|
7.25
|
16.0
|
11.7
|
0.0
|
20.0
|
1.0
|
13.0
|
3.5
|
2.0
|
9.25
|
5.5
|
0.0
|
Uganda
|
Core
|
10.0
|
1.0
|
12.0
|
3.0
|
2.0
|
4.0
|
4.0
|
0.0
|
20.0
|
1.0
|
5.0
|
1.0
|
1.0
|
1.75
|
1.5
|
0.0
|
Extended
|
10.0
|
4.0
|
3.,0
|
21.0
|
9.25
|
26.25
|
18.4
|
0.0
|
20.0
|
1.0
|
28.0
|
6.5
|
4.0
|
9.0
|
7.9
|
0.0
|
|
|
|
|
|
|
|
|
|
Home range shape
For differences in the minimum number of days for the shape index between countries, the analysis included 46 dogs (10 from Chad, 12 from Guatemala, 10 from Indonesia and 14 from Uganda) for the 10% and 20% change of the core HR; 43 dogs (nine from Chad, nine from Guatemala, 10 from Indonesia and 14 from Uganda) for the 10% change of the extended HR, as three dogs did not reach the defined threshold; and 44 dogs (nine from Chad, 11 from Guatemala, 10 from Indonesia and 14 from Uganda) for the 20% change of the extended HR, as two dogs did not reach the defined threshold. Significant differences were found for both the 10% and 20% change of the core HR (p=0.02 and p<0.01, respectively) and for the 10% and 20% change of the extended HR (p=0.02 and p<0.01, respectively).
For the analysis of the influence of the sex in the number of days for the shape index, 15 female dogs and 30 male dogs were included for the core HR, 15 female dogs and 27 male dogs for the 10% change and 15 female dogs and 28 male dogs for the 20% change of the extended HR. One dog from Uganda did not have data on sex and therefore was excluded from this analysis. Three male dogs for the 10% change and two male dogs for the 20% change never reached a minimum number of days in the extended HR and were excluded from this analysis. The effect of sex was not statistically significant neither in the core HR (10% change, p=0.95; 20% change p=0.95) nor in the extended HR (10% change, p=0.29; 20% change, p=0.49).
Minimum, maximum, median and mean number of days for each country are presented in Table 4, Supplementary Fig. S8 and Supplementary Fig. S9.
Table 4: Number of observation days required in each country to reach a constant value of percentage of change in the shape index equal or below 10% and 20% for the core and extended HR. Numbers presented here indicate the minimum (Min), maximum (Max), median, percentile 25 (Per. 25), percentile 75 (Per. 75) and mean of the number of days required amongst all dogs, as well as the number of dogs that did not reach a constant value equal or below the chosen threshold.
HR
|
Change (%)
|
Number of days
|
Number of dogs with no limits reached
|
Min
|
Max
|
Median
|
Per. 25
|
Per. 75
|
Mean
|
Chad
|
|
Core
|
10.0
|
1.0
|
43.0
|
28.5
|
1.0
|
39.75
|
22.4
|
0.0
|
20.0
|
1.0
|
42.0
|
28.5
|
1.0
|
39.75
|
22.3
|
0.0
|
Extended
|
10.0
|
1.0
|
63.0
|
23.0
|
17.0
|
30.0
|
25.8
|
1.0
|
20.0
|
1.0
|
63.0
|
17.0
|
11.0
|
23.0
|
20.6
|
1.0
|
Guatemala
|
|
Core
|
10.0
|
1.0
|
9.0
|
1.0
|
1.0
|
1.0
|
1.7
|
0.0
|
20.0
|
1.0
|
9.0
|
1.0
|
1.0
|
1.0
|
1.7
|
0.0
|
Extended
|
10.0
|
1.0
|
32.0
|
14.5
|
3.0
|
26.0
|
15.1
|
2.0
|
20.0
|
1.0
|
30.0
|
13.0
|
2.0
|
19.5
|
12.5
|
1.0
|
Indonesia
|
|
Core
|
10.0
|
1.0
|
22.0
|
1.0
|
1.0
|
1.75
|
4.5
|
0.0
|
20.0
|
1.0
|
22.0
|
1.0
|
1.0
|
1.75
|
4.4
|
0.0
|
Extended
|
10.0
|
1.0
|
14.0
|
1.0
|
1.0
|
4.75
|
3.7
|
0.0
|
20.0
|
1.0
|
10.0
|
1.0
|
1.0
|
1.0
|
2.1
|
0.0
|
Uganda
|
|
Core
|
10.0
|
1.0
|
53.0
|
1.0
|
1.0
|
1.0
|
7.4
|
0.0
|
20.0
|
1.0
|
53.0
|
1.0
|
1.0
|
1.0
|
7.4
|
0.0
|
Extended
|
10.0
|
1.0
|
56.0
|
3.0
|
1.0
|
12.0
|
11.9
|
0.0
|
20.0
|
1.0
|
50.0
|
2.0
|
1.0
|
6.5
|
6.6
|
0.0
|
Minimum required observation period considering blocks of days
When calculating the percentage of change in HR size and shape index between blocks of five and seven days, using the previously identified minimum number of days as the starting point, in most cases the percentage of change remained constantly equal or below the thresholds of 10% or 20% (Table 5). A value of zero in the number of days in Table 5 means that the percentage of change remained always equal or below the defined threshold. Yet, in some animals, a constant value equal or below the threshold took longer to be reached or was never reached (Table 5). A value higher than zero indicates the number of extra blocks required to achieve a constant percentage of change equal or below the defined threshold. While for half of the dogs no additional block of days was required (median values is zero for all, except for the 20% change threshold in the five-day block of the extended HR), for 25% of dogs (percentile 75), at least an additional block of five or seven days was needed. Furthermore, in extreme cases, up to nine (10% threshold in the five-day block of the extended HR) blocks were needed in addition.
It is also relevant to notice that some animals never reached a constant value of percentage of change equal or below the defined thresholds within the recording period (Table 5). This is mainly the case for the 10% threshold of the extended HR and 7 day-blocks for HR size, with 13 (36.1%) dogs and for the 10% threshold of the extended HR and 5 and 7 day-blocks for the shape index, both with 11 (26.2%) dogs.
Table 5: Minimum (Min), maximum (Max), median, percentile 25 (Per. 25), percentile 75 (Per. 75) and mean numbers of blocks of five and seven days required in addition to the previously identified number of days to obtain a percentage of change in HR size and shape index equal or below 10% and 20%, for the core and extended HR. Dogs without previously identified number of days or without enough remaining recorded days for a day-block comparison were excluded from this analysis. The number of dogs included are presented in column “Number of dogs included” and the number of dogs that did not reach a limit are presented in the column “Number of dogs with no limits reached”.
HR
|
Change (%)
|
Block (days)
|
Number of days
|
Number of dogs included
|
Number of dogs with no limits reached
|
Min
|
Max
|
Median
|
Per. 25
|
Per. 75
|
Mean
|
HR size
|
Core
|
10
|
5
|
0.0
|
9.0
|
0.0
|
0.0
|
1.8
|
1.4
|
45
|
3
|
20
|
5
|
0.0
|
8.0
|
0.0
|
0.0
|
1.0
|
1.0
|
45
|
0
|
Extended
|
10
|
5
|
0.0
|
5.0
|
0.0
|
0.0
|
1.0
|
0.9
|
36
|
9
|
20
|
5
|
0.0
|
8.0
|
1.0
|
0.0
|
1.5
|
1.2
|
42
|
3
|
Core
|
10
|
7
|
0.0
|
6.0
|
0.0
|
0.0
|
1.0
|
1.1
|
45
|
9
|
20
|
7
|
0.0
|
5.0
|
0.0
|
0.0
|
1.0
|
0.7
|
45
|
2
|
Extended
|
10
|
7
|
0.0
|
4.0
|
0.0
|
0.0
|
1.0
|
0.6
|
36
|
13
|
20
|
7
|
0.0
|
3.0
|
0.0
|
0.0
|
1.0
|
0.7
|
41
|
5
|
HR shape index
|
Core
|
10
|
5
|
0.0
|
1.0
|
0.0
|
0.0
|
0.0
|
0.0
|
43
|
0
|
20
|
5
|
0.0
|
0.0
|
0.0
|
0.0
|
0.0
|
0.0
|
43
|
0
|
Extended
|
10
|
5
|
0.0
|
4.0
|
0.0
|
0.0
|
0.0
|
0.5
|
42
|
11
|
20
|
5
|
0.0
|
3.0
|
0.0
|
0.0
|
0.0
|
0.1
|
45
|
3
|
Core
|
10
|
7
|
0.0
|
4.0
|
0.0
|
0.0
|
0.0
|
0.1
|
41
|
1
|
20
|
7
|
0.0
|
0.0
|
0.0
|
0.0
|
0.0
|
0.0
|
41
|
0
|
Extended
|
10
|
7
|
0.0
|
3.0
|
0.0
|
0.0
|
0.0
|
0.3
|
42
|
11
|
20
|
7
|
0.0
|
4.0
|
0.0
|
0.0
|
0.0
|
0.2
|
43
|
5
|