Large herbivore population estimates for Thanda Private Game Reserve and Mduna Royal Reserve September 2012-1 -
Large herbivore population estimates for Thanda Private Game Reserve and Mduna Royal Reserve September 2012 W. S. Matthews (Space for Elephants) November 2012 Contents I. INTRODUCTION... 2 II. III. IV. METHODS... 2 Total Aerial Count...2 Distance Sampling...3 RESULTS... 4 Known Group Estimates...5 BEST ESTIMATE OF NUMBERS...6 DRIVE COUNTS... 12 V. HABITAT ASSOCIATION... 14 VI. VII. COMPARISON WITH PREVIOUS COUNTS... 14 CONTRACEPTION PROGRAM... 15 VIII. CONCLUSIONS... 16 IX. RECOMMENDATIONS... 16 REFERENCES... 17 APPENDIX 1. DISTRIBUTION MAPS OF SIGHTINGS OF THE MOST PROMINENT LARGE HERBIVORES SEEN DURING THE THANDA PRIVATE GAME RESERVE AND MDUNA ROYAL RESERVE, SEPTEMBER 2012 CENSUS....18-2 -
I. INTRODUCTION This is the first Distance sampling based census undertaken for the most common large herbivores occurring in Thanda Private Game Reserve and Mduna Royal Reserve. The aerial census was undertaken during the beginning of September 2012, early in the season but substantial rain had fallen the preceding days before count. The overall aim of the census was to derive trends and estimates of the most common large herbivore populations in Thanda Private Game Reserve and Mduna Royal Reserve, which would be useful for management decisions and would stand as a record of abundance for future trend analyses. In addition, the survey undertaken aims to record the spatial distribution of the most important herbivores so as to gain a better sense of their habitat associations, but also to gain an understanding of the relative abundance of these herbivores related to the landscape in the Thanda Private Game Reserve and Mduna Royal Reserve. This census is based on the work undertaken in Tembe Elephant Park and Munyawana Game Reserve (Matthews 1994; Matthews 2000; Matthews 2011). The methods used; were the similar as those used in 2000, which were a refined and improved version of those used in 1994. Three methods were used to estimate numbers for each of the large herbivore populations. These were (i) total area aerial counts; (ii) transect distance sampling counts; (iii) population estimates based on the numbers of known groups such as black and white rhino. The known group estimates were derived from information gathered and reported on by reserve staff as well as other proficient people. The transect aerial census was undertaken by Tristan Parsons (pilot), Wayne Matthews (co-ordinator and recorder) with Alwyn Wentzel and Mariana Venter (observers). Thanks also go to Trish Parson of Parson s Aviation for the co-ordination of the helicopter, Digs Pascoe - Space for Elephants, staff of Thanda Private Game Reserve and specifically Alwyn Wentzel for logistical support. Thanks also to Bruce Page, Horace Olivier, Mike Toft and Mariana Venter for data ratification and general info checking. II. METHODS Aerial l Counts Total Aerial Count 1. A helicopter containing a set crew four people, pilot and recorder (front) and observers (back) was flown on pre-determined, parallel east west orientated transects arranged systematically to cover the whole census area. This resulted in 19 transects, totalling 172 km s. 2. The helicopter was flown at approximately 80 m (260 ft) above the ground and at an air speed of approximately 30 40 kts. Transects were flown, for periods of up to a maximum of 2.5 hrs. at a time, this resulted in two survey sessions as depicted in Figure 1. The hottest part of the day was avoided as far as possible, as animals tended to rest under shade in the heat of the day and as a consequence are more difficult to spot. 3. Devices were fitted to both sides of the helicopter which, when flying at a height of 80m, demarcated a predetermined distance on each side of the helicopter. All individuals sighted of all large herbivore species were recorded in this belt (transect). 4. Where large groups of eg. rhino, giraffe were spotted, the helicopter deviated from the traverse line, a total count of the group was undertaken, captured and then returned to continue the count from the point of departure, this took place 28 times. 5. All data was captured on a notebook computer using Cartalinx v 1.1 (Clark Labs, Clark University, 1999) which, when connected to the on-board GPS allowed the simultaneous collection of flight path information, animal numbers (as way points) and the number of the transect being traversed. 6. Plotting of distributions by species was done by importing the Cartalinx data into Quantum GIS 1.7.4, these in cases where the number of sightings and their distribution allowed visualisation. - 2 -
Figure 1.Census flight path for the transects for the game census. (*insert gives profile of longest transect altitude change of ~ 500m[highest point 670m and lowest 145m Distance Sampling 1. Data were collected for the distance sampling analysis at the same time as for the total count. 2. Devices were fitted to both sides of the helicopter which, when flying at a height of 80m, demarcated distance classes on each side of the traverse (transect) line. Whenever an individual or group of individuals were observed they were recorded as occurring in one of the distance sectors. 3. Animal observations recorded during the aerial census were edited and then exported directly to Distance 6 from the Microsoft Access database constructed whilst entering the data using Cartalinx. Where the number of observations allowed, density along each transect, and from this population size, was estimated using the statistical routines in Distance 6 release 2 (Thomas et al. 2009). 4. A statistically robust estimate can only be derived for species within the region of 60 sightings. Although species having less than 60 observations were analysed with Distance, these estimates should not be considered reliable (in all instances, the confidence intervals indicate this), but rather as best estimates of population size in species that have been under sampled. - 3 -
Figure 2. Weather conditions for the game census (Temperature; relative humidity; Wind speed). III. RESULTS Aerial Counts A complete aerial survey of the entire reserve took up one day to complete, taking approximately six hours. This was done in two sessions so as to allow for refuelling and breaks periods. The complete survey was not repeated. Conditions during the census were fair (figure 2). The weather was dominated by partly cloudy to cloudy, generally calm to windy (up to 15 k/h winds) conditions. The woody plant leaf flush was only partially developed, and game visibility conditions were good to very good. The count took place on the onset of a week of heavy rains in region. The altitudinal aspects of the reserve added complications to distance judgments and retaining height by the helicopter. Total Aerial Count and Distance Analysis The large herbivore population estimates from distance sampling is shown in Table 1. The distribution of sightings for the larger abundant and more important species is presented in Appendix 1. With respect to distance sampling impala, nyala, warthog and in some way giraffe, kudu, wildebeest, zebra had useable sightings or close to 60 or more sightings, and could therefore be analysed with distance (Table 1). The number of groups and the total number of animals physically counted in Thanda Private Game Reserve and Mduna Royal Reserve for 2012 are summarised in Table 7 and the introductions; predator management / removals / mortalities for the years 2011 & 2012 are summarised in Table 5 & 6. - 4 -
Table 1. 1 Large herbivore population estimates from distance sampling for Thanda Private Game Reserve and Mduna Royal Reserve 2012. Species Known Group Estimates Aerial Distance sampling estimates for 2012 Probability of detection index Rep1 95% C # (Cv) * Unreliable. # Probability of detection; the probability of recording an object (individual or cluster) in the surveyed area. (Cv - coefficient of variation of estimate [higher values more variation]). The known group estimates for black rhino is N/A, this based on the Thanda s rhino monitoring system. The number of black rhino seen from the air during the game census totalled N/A groups. The known group estimates for white rhino is N/A, this based on the Thanda s rhino monitoring system. The number of white rhino seen from the air during the game census totalled N/A groups. The known group estimates for lion is 17, this based on the Thanda s intensive monitoring system. The number of lion seen from the air during the game census totalled 9 in 2 groups. The population of buffalo was estimated at 110; based on total count refined by photographic counts and Thanda s ground monitoring system. A total of one herd was recorded, totalling 101 animals of which 6 could be classified as sub-adults and juveniles [6% of breeding population]; see Table 2. There was no lone bull groups recorded. Comparing the past count, the population seems to very similar although on the increase (Figure 5). Table 2. Summary of 2012 buffalo data, from aerial sightings, matched with photograph counts. (*- count based on best reliable photograph examined; # photo used but not totally reliable; shaded record refers to double count). 2012 Final Total Bulls Family Sub-adults & Usable Total Count group Unclassified incl Photo juv. 1 101 99-99 6 y 1 1-1 6% of fam. TOTAL 101 99-99 6 2011 1 - - 91 8% of fam. TOTAL 91 - - 7 7 Effective transect width (m) 1 Bushbuck * Bushpig n/a Giraffe 128* 54 306 / 35% 1 (0.14) 288 Grey Duiker n/a Impala 1057 589 1901 / 29% 1 (0.09) 288 Kudu 163 * 82 327 / 35% 0.92 (0.17) 265 Nyala 230 123 431 / 31% 0.60 (0.1) 119 Red Duiker n/a Reedbuck n/a Mt. Reedbuck n/a Rhino/Black n/a Rhino/White n/a Steenbok n/a Warthog 110 63 191 / 27% 1 (0.12) 288 Waterbuck * Wildebeest 254* 134 483 / 32% 1 (0.12) 288 Zebra 260* 157 431 / 25% 1 (0.11) 288 1 Effective transect width as determined by the Distance analysis - 5 -
Elephant The known group estimates for elephant is 36, this based on Thanda s ground monitoring and 2012 aerial survey. The number of elephant seen from the air during the game census totalled 33 in 3 groups. The highest number of different family units (groups) observed this year during count was 2; during contraception program 2 family units were encountered. The Maximum number of animals recorded in family groups, was 33 compared to 1 free roaming bull. Based on contraception info there would be 13 15 adult females. Summary of elephant data can be seen in table 3. The distribution of the elephant sightings is shown in Appendix Figure 20. Table 3. Summary of elephant data from contraception program and elephant sightings from October 2007 survey (Max* - refers to max number from all estimates); [% - refers to percentage of the total population composition; - refers to number of adult female contracepted; *Ayahamba Nathi Environmental Technologies report, Thanda section only]. Count Free Family Sub-adults Young Family Total roaming group & (<1-2 yrs) units Bulls Total Unclassified Count 2012 33 1 32 - - 2 Contraception 2012 - - - (13/15 ) - - 2 Photos - ID Kits 33-33 5 28 2 Local knowledge 36 2 - - - 2 Max 36 2 33 5 28 2 % 5% 91% 13% 77% Past counts 2011* 20 4-5 11 1 2010 18 - - - - - Best Estimate of o f Numbers Acceptable estimates (Table 4) for 8 species namely; wildebeest, zebra, giraffe, impala, kudu, nyala, warthog and rhino. The counts for most species, taking removals and introductions into consideration. For Distance sampling; some species, the numbers of sightings for a single repetition were marginal for a confident result, but some were analysed so as to get some indication of possible population sizes. Sample based estimates were made for six species namely, impala, nyala, kudu, warthog and zebra, wildebeest in some way. The final estimate used was based on the number of observations, best confidence levels, also taking into consideration Probability of detection, observer fatigue for the different repetitions sessions and transect data sets sighting levels. The population estimate of impala was 1057 (589 1 901) based on 54 observations (has an acceptable confidence interval; Table 1). The distribution of impala is shown in Appendix Figure 11; they are spread throughout most of reserve except for the high lying areas. With a definite trend for smaller groups in Thanda section (Table 9). The population estimated of nyala was 300 (123 431) from 49 observations, but when compared to past estimates, it possibly lower than what could be expected. Considering removals/culling figures (Tables 4 & 5), the overall trends could very well be an indication of a decline, supported by the trend of predator management / baits used trends on decline for this species. The distribution (Appendix Figure 12) of nyala is spread throughout, although the bulk of sightings coming from Thanda main (table 9). The population estimate for kudu was 160 (82 327) and was based on a sample size of 21 observations, below what is required for reliable statistical estimation and as also reflected by confidence value. The distribution of kudu is shown in Appendix Figure 13; sightings seem to be spread though out reserve with a definite trend for bigger groups in Mduna section (Table 9). The giraffe population was estimated at 84 from 26 observations; it was considered that the total count was relatively reliable. The distribution of giraffe is shown in Appendix Figure 14, and with - 6 -
bulk of sightings coming from Thanda main, although number wise an even split (table 9), as again a trend of smaller groups on Thanda main. The warthog population is estimated at 110 (63 191), this was determined from 35 observations, (has relatively good Distance confidence intervals - though below min sightings needed for a reliable distance estimate). This can be considered a reasonable estimate. The distribution (Appendix Figure 15) of warthog was widely spread throughout most of reserve except for the high lying areas, and a distinct prevalence in the link area between Thanda main and Mduna. The wildebeest population is estimated at 254 (134 483), which was determined from 32 observations, below what is required for reliable statistical estimation and as also reflected by confidence value. As with giraffe, wildebeest are also found in the more open areas [probability of detection=1], and thus the possibility exists that the statistical result could be bias, because of this, it was felt to lean on the side of caution with the estimate and not possibly overestimate the population size. The distribution of wildebeest is shown in Appendix Figure 16, was widely spread, with concentrations in central areas of the Reserve and, locally absent from the high lying areas Northern Zululand Sourveld. Again, a distinct trend of smaller groups on Thanda main (Table 9). Table 4. 4 Final estimates for Thanda Private Game Reserve and Mduna Royal Reserve 2012. Estimation Method: 1 Known Group, 2 Total Area Count, 3 Distance Sample, 4 Informed Guess, 5 Field encounter rates, 6-Informed guess based on past estimates, 7-based on photograph count. (* unreliable; ~ approximate) Total Count Distance Sample Known Final Estimate Group Bushbuck 8 8 Buffalo 99 /101 7 ~110 ~110 Giraffe 84 54 306 / 35% 84 2 Elephant 33 36 36 Grey Duiker D 7 7 Red Duiker 3 3 Impala 694 589 1901 / 29% 1057 Kudu 99 82 327 / 35% 160 Nyala 109 123 431 / 31% 300 Reedbuck 13 13 Mt. Reedbuck 1 1 Warthog 75 63 191 / 27% 110 Waterbuck 21 21 Wildebeest 188 134 483 / 32% 254 Zebra 171 157 431 / 25% 260 Rhino/Black Rhino/White Hyena 1 1 B B jackal 1 1 Leopard - 1 Lion 9 17 17 1 Cheetah 2 3 3 1 Wilddog - 14 14 1 The zebra population is estimated at 260(157 431), which was from 39 observations, (has a relatively good Distance confidence interval - the best of all species analysed even though below min sightings needed for a reliable distance estimate). Also, similar to wildebeest, found associated with the more open areas [probability of detection=1]. The distribution of zebra is reasonably wide spread throughout the Reserve (Appendix Figure 18), similar patterns to wildebeest, again locally absent from the high lying areas Northern Zululand Sourveld. Again, a distinct trend of smaller groups on Thanda main (Table 9). - 7 -
The population size of reedbuck was estimated at 13 from a sample of 8 sightings, below what is required for reliable statistical estimation. All of the reedbuck was found in the Mduna section(figure 17), with only one sighting of mountain reedbuck, as could be expected coming from the tops of the hills of the Mduna section. The sightings of red duiker and other of the smaller antelope species (eg. grey duiker, bushbuck) would be low as a result of counting technique, this was expected, a result of higher flying levels and broader transects. The true population size of these smaller discreet species could be much larger. The distribution the few sightings of these species can be seen in Appendix Figure 9. Predator management and predator related mortalities were a major component of the recorded mortalities for the reserve (Figure3; table 6). Based on the reserves mortality/carcass monitoring; predator related mortalities were in fact responsible for more than 80% of all recorded mortalities over a three year period. Table 5. Introductions records for Thanda Private Game Reserve and Mduna Royal Reserve 2009 2012. Species 2009 2010 2011 2012 Total Giraffe Kudu Nyala 13 Impala 150 662 Wildebeest 26 Black rhino * White rhino * Zebra 19 # White rhino introduced into Mduna Royal * - Best info at time of completion of report 250 200 150 100 50 Pred. Baits Culling Mort. Removal # 0 2010 2011 2012 Figure 3. Summary of Removals, mortalities and predator management trends for Thanda Private Game Reserve and Mduna Royal Reserve, for 2010 to 2012. - 8 -
Table 6. Removals, mortalities and predator management records for Thanda Private Game Reserve and Mduna Royal Reserve. (- No mortality records; # denotes live capture; * - poached.) 2010 2011 2012 Total for past Species Removal Mort. Culling Baits Pred. Total Removal Mort. Culling Baits Pred Total Removal Mort. Culling Baits Pred Total 3 years # etc # etc # etc Bushbuck - Buffalo 1 5 6 9 9 1 3 4 19 Bushpig - Porcupine 1 1 1 Giraffe 1 1 2 1 1 2 1 1 5 Grey Duiker 8 8 4 4 1 1 13 Red Duiker 1* 21 22 1 1 1 3 4 27 Kudu 2 12 14 1 3 4 1 1 2 20 Steenbok 3 3 3 Impala 2* 4 21 51 78 5* 1 9 44 56 101 1 2 104 12 119 298 Nyala 1* 1 35 37 1* 29 30 3 3 70 Reedbuck 3 3 4 4 7 Mt Reedbuck - Warthog 1 3 21 25 2* 12 20 34 1 20 1 22 81 Waterbuck - Wildebeest 1 2 20 23 2 25 27 1 4 9 14 64 Zebra 4 4 1 1 2 1 1 7 Rhino/Black 1 1 1* 1 2 Rhino/White 4 1 5 2* + 2 4 9 lion 1* 1 1* 1 2 Wildog 2 2 2* 2 4 Leopard - Hyena 2* 1 3 2 2 5 Total 12 15-26 185 237 11 6 9 59 153 224 5 4 3 129 35 173 636-9 -
Table 7. Summary of large herbivore total counts (minimum number), for Thanda Private Game Reserve and Mduna Royal Reserve 2012. (2012 count based on a new systematic flight grid made up of 18 transects). {red lower; green higher than preceding year s count}; * Correction based on photo analysis. # Refers Max of No counted / sightings. Species Black rhino Road Census 2003 03 Thanda main No. No. Groups Counted Road Census 2004 04 Thanda main No. No. Groups Counted Road Census 2005 05 Thanda main No. No. Groups Counted Road Census 2006 06 Thanda main No. No. Groups Counted Road Census 2011 Thanda main #No. #No. Groups Counted Aerial Census 2012 Thanda main Aerial Census 2012 Thanda - Mduna No. No. % 2 No. No. Groups Counted Groups Counted Buffalo 1 18 1 29 2 25 1 91 1 99 100 1 99 Bushbuck 1 1 1 1 2 2 1 1 0 0 1 1 25 4 4 Bushpig 0 0 0 0 0 0 C Reedbuck 3 4 2 2 1 2 5 6 1 2 0 0 0 8 13 M Reedbuck 1 2 0 0 0 0 0 1 1 Elephant 2 11 1 11 1 2 1 2 1 21 49 3 33 Giraffe 2 6 10 32 4 21 5 11 21 48 57 26 84 Grey Duiker 10 11 7 7 4 4 8 9 0 0 2 2 29 7 7 Impala 29 169 74 457 68 519 108 631 32 223 43 452 65 54 694 Kudu 9 30 10 40 14 26 11 34 13 42 42 21 99 Nyala 80 193 49 101 80 191 65 170 43 106 39 85 78 49 109 Red Duiker 23 27 7 7 13 15 15 15 3 3 1 1 33 3 3 Steenbuck 12 13 3 3 10 11 8 11 2 3 1 1 100 1 1 Warthog 24 50 29 54 43 119 67 171 18 34 28 129 79 35 75 Waterbuck 1 4 2 8 6 15 0 0 0 0 0 6 21 White rhino Wildebeest 1 11 15 176 14 103 19 109 20 163 28 59 69 32 188 Zebra 8 144 17 76 12 66 21 95 26 79 46 39 171 Cheetah 1 1 1 2 100 1 2 Lion 1 3 2 9 100 2 9 jackal 1 1 1 1 1 1 Wild dog 0 0 0 0 0 0 Hyena 0 0 1 1 1 1 Leopard 0 0 0 0 0 0 2 % that Thanda main plays to overall Total - 10 -
Table 8. Summary of large herbivore estimates, for Thanda Private Game Reserve and Mduna Royal Reserve 2012. (2012 count based on a new systematic flight grid made up of 18 transects). (* unreliable; ~ approximate) Road Census Road Census Road Census Road Census Road Census estimate 2003 03 estimate 2004 04 estimate 2005 05 estimate 2006 06 estimate 2011 Thanda main Thanda main Thanda main Thanda main Thanda main Aerial Census estimate 2012 Thanda main 3 Aerial Census estimate 2012 Thanda / Mduna Black rhino Buffalo 20 33 39 50 87 101 101 Bushbuck - 10 10 10 0 >1* >4 Bushpig - - - - 0 - - C Reedbuck 35 35 30 21 10 >1* >13 M Reedbuck - 20 20 20-25 0 0 >1 Elephant 12 11 11 13 20 21 36 Giraffe 20 23 25 39 60 48 84 Grey Duiker 63 31 23 29 0 >2* >7 Impala 648 1047 1239 1102 682 688 1057 Kudu 40-60 62 72 47 66 69 163 Nyala 987 337 590 388 236 230* 300* Red Duiker 227 37 52 54 19 >1* >3 Steenbuck 40 21 24 17 15 >1* >1 Warthog 180 132 297 289 116 90 110 Waterbuck 14 16 23 19 19 0 >21 White rhino Wildebeest 87 129 157 166 271 174 254 Zebra 56 95 124 121 180 120 260 Cheetah 2 3 3 Lion 11 9 9 jackal - - - Wild dog 6 14 14 Hyena - - - Leopard - - - 3 Based on percentage ration between the 2 sections see Table 7. - 11 -
Table 9. Summary of 2012 survey data, from aerial sightings, showing break down of the two sections IV. Drive counts Figure 4. Map showing road drive routes used in a portion of Thanda Game Reserve (Full layout of road drives not available presently). - 12 -
In May 2012, a rotation of five drive routes where initiated to collect game sighting data which could be used to assess game abundances, structures and trends over the seasons. To-date each drive has been repeated between 4 6 times. Only one drive route is currently present on Mduna section, which has resulted in a low effort in this area. The summery of the results of this data collection, which took place from May to September 2012, is presented in Tables 10 & 11. In undertaking this drive count, a total distance covered was 8 077 km; and the total time spent on the drivers was approximately 28.7 hours. Species overall encounter rates (indication of abundance levels), show similar patterns to the overall aerial survey results and even though the effort was less on the Mduna section, the encounter rate of animals is very similar to Thanda section (table 10). With continued repeats of drive counts more rigorous interrogation of data can be undertaken. Table 10. Summary of 2012 drive count survey data, presenting break down of the two sections. [Sighting index (#/km) gives an indication of sightings per unit effort] Table 11. Summary of sex ratios of species based on 2012 drive count data. Adult males % Adult female % Unidentified Adults % Yearling % Bushbuck 1 25% 3 75% 1 25% Buffalo 3 Giraffe 42 66% 17 27% 5 7.8% 3 5% Impala 80 13% 506 81% 42 6.7% 14 2% Kudu 14 22% 51 78% 3 5% Nyala 63 41% 89 59% 8 5% Waterbuck 3 43% 3 43% 1 14.3% Warthog 6 12% 15 29% 30 58.8% 4 8% Wildebeest 60 31% 75 39% 56 29.3% 9 5% Zebra 23 16% 48 33% 74 51.0% 1 1% - 13 -
V. Habitat Association The habitat association trends for the survey results of 2012, was not undertaken at present, as there is only one year s data. Only, possibly, one broad association to a veldtype can be made; and that was that many species were found locally absent in the high lying areas Northern Zululand Sourveld. More reliable selection association / indices will emerge with continued surveys, collecting this type of information. VI. Comparison With Previous Counts Total Numbers Trends The species estimate trend / total count trends for most species from 2003 to 2012 are graphically represented in Figures 5, 6 & 7. As this is the first year of collated data, confident interpretations of long-term trends at this stage are of partial value. Taking the introductions, predator management and removals/mortalities into consideration, most species showing a decrease or stable trend; reedbuck reaching low levels after a long continued downward trend; while only buffalo and elephant showing an expected increase on the estimates. Wildebeest and Zebra showed a significant higher than expected estimate for 2011, but this may not be a real in numbers increase but a result of an artefact of the 2011 counting conditions; this will only become clear when repeat surveys are undertaken. This decrease can partly be ascribed to a combination of factors namely (i) a real decrease in numbers and (ii) counting technique and (iii) visibility/vegetation cover. This year s decrease may be a real change in numbers; a result of management interventions (eg predators/ management) and the precision of past estimates /data and in part an artefact of the 2012 counting technique. Follow up surveys results should make this clear. 250 200 GIRAFFE C. REEDBUCK WILDEBEEST ZEBRA BUFFALO Number 150 100 50 0 2003 2004 2005 Year 2006 2011 2012* Figure 5. Summary of selected large herbivore, of more visible species (where probability of detection ~1), estimates trends for 2001 to 2012. - 14 -
70 60 ELEPHANT KUDU WATERBUCK 50 Number 40 30 20 10 0 2003 2004 2005 Year 2006 2011 2012* Figure 6. Summary of selected, relative lower abundance, herbivores estimates trends for 2001 to 2012. 1400 1200 1000 Number 800 600 400 200 0 2003 2004 2005 2006 2011 2012* IMPALA 648 1047 1239 1102 682 688 NYALA 987 337 590 388 236 230 WARTHOG 180 132 297 289 116 90 Figure 7. Summary of nyala, impala and warthog estimate trends for 2001 to 2012. VII. Contraception program The total number of family units found, number of young and adult females darted during the two contraception operations is shown in Table 2. No information on adult bulls or group sizes were gained from this operation. During the operation no count of animals was attempted. Table 12. Summary of September and November 2012 operation, breeding groups found and darted. Total animals Females Juvenile Young Sub-adults Group Date in group Darted (<1-2 2 yrs) (2<4 yrs) & Time [contracepted] Unclassified Notes 1 14 Sep - 6 - - - Mduna 2 14 Sep - 7 - - - Thanda Total 13 1hr 9 16 Nov - 7 - - - Mduna 10 16 Nov - 8 - - - Thanda Total 15 1.5hr - 15 -
Figure 8. The inception elephant contraception program of 2012; Vet Dr Mike Toft Field loading of marker darts with vaccine. VIII. Conclusions Census conditions were on the whole acceptable. The two methods employed namely, total area counts and distance based sample estimates, produced defendable population estimates. Most species count results did not show the expected increasing trends, when comparing introductions and mortalities numbers. Predator management and predator related mortalities were a major component of recorded mortalities. Predator related mortalities were >80% of all recorded mortalities over a three year period. Many species, like impala, wildebeest showed a strong tendency on Thanda main, to be found in smallish groups, below what would be the expected average clustering size. This was not the case for the Thanda linkage area or Mduna. Wildebeest and Zebra showed a significant lower than expected estimate, based on 2011 result, but this may a result of an artefact of the 2011 counting conditions, this will only become clear when repeat surveys are undertaken. The sightings of the smaller antelope species will be low as a result of counting technique. More confident estimates will be produced if distance sampling counts continue. Drive counts have potential to generate an effective database from which overall monitoring over time can be used for management decision, but effort needs to be increased, overall and specifically in the Mduna section. IX. Recommendations It was felt that the 2012 census effort was a success, although some recommendations can be made and should be incorporated into the follow-up census programs. 1. It would be meaningful for Thanda Private Game Reserve and Mduna Royal Reserve, that followup census are undertaken using the methods described and reported on here. 2. Try to complete census at a flight speed of around 30kts, taking topography and wind into consideration. 3. Comparative analyses of the trends of the more important species should continue to be undertaken so as to have long-term defendable trends, especially in the case of impala, nyala, warthog, wildebeest, zebra and reedbuck. - 16 -
4. Predator impact / management needs to be evaluated, in terms of effect it may be having on prey base. Carcass monitoring and categorization needs to be treated as key information. 5. Drive counts effort needs to be increased so that a more rigorous interrogation of data can be undertaken. 6. Other counts to supplement count for smaller herbivores, such as bushbuck and grey duiker. 7. Reserve data; an overall increased in thoroughness of data consistency of collection and recording is required. Reserve data evaluation and archival needs to be more rigorous and consistent. References Matthews, W.S. 1994 Aerial surveys for Tembe elephant Park. Winter 1994. 1 11. KwaZulu Department of nature conservation unpublished report. Matthews, W.S. 2000. Large herbivore population estimates for Tembe Elephant Park: October 2000. KwaZulu- Natal Nature Conservation Services unpublished report. 1 15. Matthews, W.S. 2011. Large herbivore population estimates for Munyawana Game Reserve. September 2011. WSM Eco Services, Nelspruit; unpublished report. 1 28. Thomas L., Laake, J.L., Rexstad, E., Strindberg, S., Marques, F.F.C., Buckland, S.T., Borchers, D.L., Anderson, D.R., Burnham, K.P., Burt, M.L., Hedley, S.L., Pollard, J.H., Bishop, J.R.B. and Marques, T.A. 2009. Distance 6.0. Release 2. Research Unit for Wildlife Population Assessment, University of St. Andrews, UK. http://www.ruwpa.st-and.ac.uk/distance/ Van Rooyen, N. 2012. Habitat evaluation, vegetation and wildlife management of THANDA and MDUNA. Part 1: Vegetation. Unpublished Report. Ekotrust CC. - 17 -
Appendix 1. Distribution Maps of sightings of the most prominent large herbivores seen during the Thanda Private Game Reserve and Mduna Royal Reserve,, September 2012 census. 4 (The symbol size is an indication of group size). Figure 9. Distribution and group sizes of the low frequency species such bushbuck (Black) red duiker (red arrows with green) and grey duiker(red arrows with grey). 4 Map background represents vegetation types as distinguished by Van Rooyen 2012. - 18 -
Figure 10. Distribution of. Figure 11. Distribution and herd sizes of impala. - 19 -
Figure 12. Distribution and group sizes of nyala Figure 13. Distribution and group sizes of kudu. - 20 -
Figure 14. Distribution and group sizes of Giraffe. Figure 15. Distribution and group sizes of warthog - 21 -
Figure 16. Distribution and group sizes of wildebeest. Figure 17. Distribution and group sizes of reedbuck. - 22 -
Figure. 18. 1 Distribution and group sizes of zebra. Figure. 19 1.Distribution of predator sightings. - 23 -
Figure. 20. Distribution of elephant sightings. - 24 -