Trophy Hunting and Sustainability: Temporal Dynamics in Trophy Quality and Harvesting Patterns of Wild Herbivores in a Tropical Semi-Arid Savanna Ecosystem

The selective nature of trophy hunting may cause changes in desirable phenotypic traits in harvested species. A decline in trophy size of preferred species may reduce hunting destination competitiveness thus compromising the sustainability of trophy hunting as a conservation tool. We explored the trophy quality and trends in harvesting patterns (i.e., 2004–2015) of Cape buffalo (Syncerus caffer), African elephant (Loxodonta africana), greater kudu (Tragelaphus strepsiceros) and sable (Hippotragus niger) in Matetsi Safari Area, northwest Zimbabwe. We used long-term data on horn and tusk size, age, quota size allocation and offtake levels of selected species. To analyse the effect of year, area and age on the trophy size, quota size and offtake levels, we used linear mixed models. One sample t-test was used to compare observed trophy size with Safari Club International (SCI) minimum score. Trophy sizes for Cape buffalo and African elephant were below the SCI minimum score. Greater kudu trophy sizes were within the minimum score threshold whereas sable trophy sizes were above the SCI minimum score between 2004 and 2015. Age at harvest for Cape buffalo, kudu and sable increased whilst that of elephant remained constant between 2004 and 2015. Quota size allocated for buffalo and the corresponding offtake levels declined over time. Offtake levels of African elephant and Greater kudu declined whilst the quota size did not change between 2004 and 2015. The quota size for sable increased whilst the offtake levels fluctuated without changing for the period 2004–2015. The trophy size and harvesting patterns in these species pose a conservation and management dilemma on the sustainability of trophy hunting in this area. We recommend: (1) temporal and spatial rotational resting of hunting areas to create refuge to improve trophy quality and maintenance of genetic diversity, and (2) introduction of variable trophy fee pricing system based on trophy size.


Introduction
Wildlife conservation is characterised by proprietorship and pricing systems within the biological parameters that limit sustainable utilization [1,2]. This is opposed to wildlife preservation approaches which promote restraint in the harvest and consumption of wildlife species and their products [3]. In sub-Saharan Africa, wildlife conservation in protected areas [4] is substantially supported by revenue generated through sustainable harvesting of wildlife species through trophy hunting in Category VI protected areas [5,6]. Trophy hunting refers to hunting by paying tourists, typically with the objective of selecting individuals with exceptional phenotypic traits (e.g., large horns, tusks, body size, mane or skull length) and usually in the company of a professional hunting guide [7]. Though there has been perpetual debate and polarity on the sustainability of trophy hunting in most sub-Saharan African countries [8], it is still considered as one sustainable way of supporting conservation in African countries endowed with abundant wildlife species [9].
In this study, sustainability refers to the ability of trophy hunting to support, sustain and ensure persistence of wildlife populations without compromising their abundance and diversity over a long-term within the framework of intergenerational equity [10,11]. In inaccessible, remote and marginalised areas lacking infrastructure, attractive scenery or high densities of charismatic and viewable wildlife species, trophy hunting is considered more suitable and feasible alternative for revenue generation over other forms of tourism (i.e., ecotourism, photographic tourism, enclave tourism [12]) [13,14]. It is also becoming evident that trophy hunting may provide revenue generation for conservation opportunities in countries where other forms of tourism may not be suitable due political instability [15] and negative media framing [16].
Central to the controversy of trophy hunting is the continual decline and possible expiation of wildlife populations in most sub-Saharan countries [17][18][19]. Historically, unregulated hunting in some continents led to the extinction of some wildlife species through what has been referred to as the global blitzkrieg (overkill) hypothesis [20,21], whilst the African continent was to some extent spared from this unprecedented loss of species due to over harvesting and illegal hunting activities [22]. In recent times, declines in wildlife populations globally (including Africa) have been associated with among others, illegal hunting [23][24][25][26][27][28], over harvesting [29][30][31], droughts [32][33][34] and fragmented and weak hunting policies that regulate harvesting of wildlife species [35]. However, trophy hunting uses a quota system approach that promotes sustainable off-takes by harvesting small portions of the natural population growth rates which arguably falls within the compensatory mortality range and has a negligible impact on overall ecology of wildlife species [36,37].
A quota refers to the number of individuals of a particular animal species that is legally allocated or prescribed for harvesting per year for a particular area [38]. The quota system used is based on ecological theory, i.e., maximum sustainable yield (MSY), set in such a way that offtake levels are always below the growth rate of the target species at any given time [39,40]. Accordingly, trophy hunting is meant to remove only a few individuals, mostly those that have passed their prime reproductive time and as such should not compromise viability of wildlife species [41][42][43].
The size of a quota allocated for trophy hunting is mostly influenced by several factors such as the population size [39], trophy size, hunting success [44], age at harvest [45], habitat management and whether or not the populations are shared by two or more management regimes (e.g. in the context of KAZA TFCA). The frequency at which these factors are monitored and analysed to inform the quota setting process as part of the adaptive management process is often low and not consistent [41,46]. Fragmented monitoring programs of these parameters are mostly a result of the long-term costs associated with their monitoring over time, thus compromising the effectiveness of the process [47]. Moreover, the use of MSY tends to be problematic as it has been developed and mostly utilised in aquatic ecosystems where severe declines in important fish catches have been witnessed [48,49]. Though the MSY concept is theoretically sound, its applicability in reality is marred with several challenges such as politics, fixed quotas, shared populations which may ultimately cause declines in wildlife species [50,51].
The sustainability of trophy hunting in Category VI protected areas in most southern African countries is increasingly subjected to scrutiny both from an ecological and ethical perspective [52]. It is becoming evident that there are some negative effects of trophy hunting on the phenotypic traits and population dynamic of hunted species. Some studies have shown that selective harvesting related to trophy hunting may result in the loss of the more desirable phenotypic traits (i.e., horn or tusk size) with increasing hunting pressure [53]. Despite these observations, there are few studies reporting on the decline of trophy size in hunting destinations in southern Africa, e.g., Zambia [54], northwest Zimbabwe [55], and South Africa [56,57]. However, the declines in horn size cannot be attributed solely to selective hunting pressure let alone inbreeding depression but a combination of these with some environmental factors [58]. Nonetheless, little attention has been given to establish the relationship between observed trophy size and the standard trophy size of harvested species [44].
In Zimbabwe, trophy hunting mainly occurs in safari areas, communal areas and private areas [59]. The Zimbabwe Parks and Wildlife Management Authority administers a participatory quota setting system with the concerned stakeholders (i.e., private land owners, communal areas representatives and private concessionaires in state owned safari areas) as a way of controlling the offtake levels through trophy hunting. The north-western side of Zimbabwe is covered by Matetsi Safari Area, one of the prime hunting areas known to have conservative and considerably low quota allocations for some wildlife species [51]. These low quotas (< 5% of the target population size) are believed to have a considerable effect on the population size let alone the horn or tusk size of targeted species [60].
Conservationists argue that there is much uncertainty over the sustainability of offtake rates and their potential impacts on wildlife populations. For instance, the United States of America has taken bold steps in banning import of ivory and related products especially from Kenya and Zimbabwe since 2014. Coincidentally, some commercial passenger and cargo airlines have also put in place an embargo on the transportation of trophies of legally and sustainably hunted species [9]. These embargoes have been worsened by the negative and emotive media framing of trophy hunting in Zimbabwe following the controversial killing of 'Cecil' the lion (Panthera leo) by an American hunting tourist near Hwange National Park [8]. This negative media framing of a country may reduce its attractiveness as a destination, which result in low offtake levels of species thus reducing revenue generation from trophy hunting [16]. Furthermore, considering the negative media framing of Zimbabwe during the period of political instability and economic decline, 2000-2008 [16], as well as restrictive policy on trophy imports [9], international trophy hunters may avoid Zimbabwe as a hunting destination thus reducing the trophy hunting offtake levels in hunting areas compared to the period of political inclusiveness and economic recovery (2009)(2010)(2011)(2012)(2013)(2014)(2015).
Although several studies have been done on trophy hunting of lions and leopards (Panthera pandus) [46,[61][62][63][64], few studies have explored on trophy size related issues on large wild herbivores in southern Africa [44, [55][56][57]. In this study, we explored the temporal dynamics in trophy quality and harvesting patterns of four selected wild herbivores, Cape buffalo (Syncerus caffer), African elephant (Loxodonta africana) and mid-sized herbivores, greater kudu (Tragelaphus strepsiceros) and sable (Hippotragus niger) in a semi arid tropical ecosystem, Matetsi Safari Area, a hunting complex within the Kavango Zambezi Transfrontier Conservation Area (KAZA TFCA), northwest of Zimbabwe. We tested three hypotheses, (1) selective harvesting through trophy hunting may result in reduced horn or tusk size and age at harvest of selected wild herbivores with the passage of time, (2) sustainable utilization management programs may reduce the quota size allocated for selected wild herbivores and their offtake levels over time commensurate with the population and trophy size trends in different hunting areas, and (3) economic status of the country between the period 2004-2015 would have an effect on quota size and offtake levels in Matetsi Safari Area, Zimbabwe.

Study Area
The study was conducted in an unfenced protected area network, Matetsi Safari Area that covers approximately 3,000 km 2 , northwest of Zimbabwe (Fig 1). Matetsi Safari Area is part of the Kavango Zambezi Transfrontier Conservation Area (KAZA TFCA) which is shared between Angola, Botswana, Namibia, Zambia and Zimbabwe established in 2011 [65,66]. In Zimbabwe, Protected Areas Category VI is referred to as Safari Areas. These areas occur mainly surrounding National Parks and are managed mainly for the sustainable use of natural ecosystems and as part of a buffer zone to cushion National Parks from human disturbances [4]. Matetsi Safari Area is divided into seven hunting management blocks called Units ( Table 1).
The southern block (i.e., comprise of Unit 1-5) is boarded by Hwange National Park to the southern part, north-eastern side with private and communal areas whereas the western side is mostly Kazuma Pan National Park and Forestry Area Hunting block. However, the northern block (i.e. Unit 6 and 7), are sandwiched by protected areas, Zambezi National Park to the eastern side, and to the western side are Chobe National Park, Botswana and Forestry Area to the south. Trophy hunting has been the sole land use option for Matetsi Safari Area for more than 37 years [67]. The main soil types on sites are lithosols and regosols occurring on Karoo volcanic and Kalahari geological formations, respectively [68]. The lithosols are dominated by Colophospermum mopane and Terminalia species. [69] whilst Baikiaea plurijuga, which occurs in association with Pterocarpus angolensis and Guibortia coleosperma dominate on the regosols [68]. Some of the common wildlife species in the study area include: herbivores (African buffalo, Burchell's zebra (Equus quagga), elephant, giraffe (Giraffa camelopardalis), greater kudu, impala (Aepyceros melampus), reedbuck (Redunca arundinum), sable antelope, warthog (Phacochoerus aethiopicus) and waterbuck (Kobus ellipsiprymnus), wildebeest (Connochaetes taurinus)) and carnivores (leopard, Lion, Hyena (Crocuta crocuta)).
Study species. The following wild herbivores were used as study species for this study: mega herbivores, Cape buffalo, elephant and medium-sized herbivores, greater kudu and sable. These four species were selected on the basis that they are amongst the most commonly hunted herbivores in southern Africa [13,44,71]. In addition, complete records on trophy size, quota allocations and utilization levels for the species were readily available for the period 2004-2015 at Matetsi Safari Area headquarters. The densities of these species in this area have been documented by Crosmary, Côté [51] as: Cape buffalo (1.4 individuals per km 2 ), African elephant (0.7 individuals per km 2 ), greater kudu (1.4 individuals per km 2 ) and sable (0.7 individuals per km 2 ) for the period 1995-2010. During this same period, the average harvest rates (i.e., number of individuals harvested per year divided by the total population estimate for that year [51]) was 1.7 ± 1.2% for the four species.

Data Collection
We collected long-term data for the six hunting units (i.e. Unit 1-6, see Table 1), from Matetsi Safari Area headquarters on; (1) trophy size and age at harvest data for the period 2004-2015, and (2) annual quotas allocation and offtake levels for African elephant, Cape buffalo, greater kudu and sable for the period 2004-2015 for Unit 1-6 (S1 and S2 Files). All trophy size measurements for the four species were done by parks rangers following the Safari Club International (SCI) scoring system (http://www.scirecordbook.org/docs/methods). The associated data on age at harvest for the four wild herbivores was estimated by parks rangers using dentition and jaws [72][73][74] as part of their monitoring routine at Matetsi Safari Area. We considered the number of animals harvested off an allocated quota for each year as the offtake level. We further determined offtake growth rate per species, i.e., mean annual change in offtake size, which was calculated using the following formulae after Rist, Milner-Gulland [75]: where h t is the size of the total offtake in year t: We expected that the offtake rates would be analogous to population size and happens to be positively correlated [76], thus can be used as proxy or index for population estimates of harvested species. Accordingly, we used data of annual utilization of the quota allocation for each hunting unit for the period 2004-2015 except for the African elephant with an incomplete data set so we used data for 2005-2015.
Data analysis. Data on trophy size, age at harvest, quota size and offtake levels were tested for normality and equality of variance to using Shapiro Wilk test and Levene's test respectively to ascertain if the normality assumptions were being satisfied. All data on explanatory variables, i.e., trophy size, age at harvest, quota size and offtake level were found to conform to the normality assumptions. We grouped data on quota size and offtake levels into two time intervals based on the temporal economic status: (a) period of land reform, hyper inflation and policy changes, 2004-2009, and (b) period of political stability, deflation and economic recovery, 2009-2015, in Zimbabwe. All the data were analysed separately for each species.
First, we computed a simple linear regression to assess the temporal trends in the trophy size and age at harvest for the four herbivores. A linear mixed model (LMM) was used to analyze the variation in trophy size, age at harvest and offtake levels trends for the six hunting units for the period 2004-2015. The model parameters included the trophy size as the dependant variable whilst year and area were fixed variables and the age at harvest being the covariate. We further conducted a one-sample t-test to ascertain if the observed trophy size differed from the SCI minimum score for greater kudu (121 inches), sable (96 inches), Cape buffalo (101 inches) and African elephant (90 pounds) (http://www.scirecordbook.org). Second, we used Analysis of Variance (ANOVA) to assess the spatial variation in quota size and trophy size of the selected herbivores across the six hunting units. Significant effects were followed by pair-wise contrasts using sequential Bonferroni post hoc adjusted significance (p < 0.05). Third, to establish the effect of temporal economic status on harvesting patterns, we computed an independent t-test to compare the overall quota size and offtake levels for the two time periods, i.e., 2004-2008 and 2009-2015. We conducted all statistical analyses in IBM SPSS 20 software package (IMB, New York, USA) at 5% level of significance.

Trophy Size and Age at Harvest Patterns
The total number of harvested individuals for the four selected wild herbivores was: We found no significant trends in the trophy size of Cape buffalo, Greater kudu and sable (p > 0.05) for the period 2004-2015 (Fig 2A, 2C and 2D). However, the trophy size of African elephant declined significantly (β ± SE: -1.03 ± 0.45, t = -2.29, p = 0.023) for the period 2004-2015 (Fig 2B). The effect of area was not significant though its interaction with year and age at harvest was significant for the four species during the same period ( Table 2). The temporal patterns on age at harvest for Cape buffalo, greater kudu and sable recorded for the period 2004-2015 were significant (Table 3). On the contrary, African elephant age at harvest did not change over time for the period 2005-2015 (R 2 = 0.01, β ± SE: -0.26 ± 0.17, t = -1.58, p = 0.115; Table 3, Fig 3).
For the period 2004-2015, only greater kudu mean age at harvest and trophy size did not vary with the hunting Unit, i.e., age at harvest (F (5, 560) = 1.35, p = 0.385) and trophy size (F (5, 560) = 0.24, p = 0.859) ( Table 4). Bonferroni post hoc test however showed that the trophy size and age at harvest for individuals in Unit 6 were higher than the other Units except for Cape buffalo (Table 4).

Effect of Temporal Economic Status Harvesting Regime Patterns
The period 2004-2015 was characterized by significant changes in the offtake growth rate for Cape buffalo (F (11, 60) = 2.01, p = 0.043). However, we recorded no difference in the quota size (t (70) Fig 6A). Similarly, there were no differences in the quota size (t (70) Fig 6C). The quota size allocated for sable for the period 2004-2008 were lower than those for the 2009-2015 period (t (70) = -4.77, p < 0.000). However, we did not record any difference in the offtake levels of sable between the same two periods (t (70) = 0.52, p = 0.602, Fig 6D).  size of wild herbivores over time have been reported in South Africa [56,57] and Tanzania [44]. Declines in trophy size over time due to selective harvesting could be attributed to phenotypic plasticity [77] that may result due to a decline in abundance of big tuskers and individuals with big horns or tusks as these are mostly selected by hunters. However, our study examines data for a fairly short period (i.e., 11 years) and as such we may not attribute the observed changes to a possible genetic effect of selective pressures that favour expression of small horns or tusks [78][79][80].

Fig 2. Temporal trend in mean trophy size for the harvested wild herbivores, (a) Cape buffalo, (b) African elephant, (c) greater kudu, and (d) sable for the period 2004-2015 in Matetsi Safari
Our results showed that African elephants' age at harvest did not change over time though their trophy size declined. However, in this study for the period 2004-2015 we found sable age at harvest to have increased significantly contrary to the observations for the period 1979-2005 in the same area [55]. On the contrary, the trophy sizes for Cape buffalo and greater kudu have not changed for the period 2004-2015 in Matetsi Safari Area as reported in some countries, e.g., Tanzania [44] and South Africa [56,57]. Most of the documented studies done on African elephant relates to illegal hunting effects on the tusk size instead of trophy hunting related issues and as such there were no comparative studies [81]. However, we note that most illegal hunting of elephants target the large tuskers and as such could have the same effect of trophy hunting.
Variations in trophy size and age at harvest could be as a result of several factors including (1) use of the fixed quota system that reduces the density or availability of old trophy individuals with the requisite trophy sizes, (2) lack of consistent age based trophy harvesting policy that penalises the harvesting of young individuals [46,62,64], (3) habitat quality heterogeneity that affect horn development and growth of trophy species [82,83], and (4) possible effects of illegal harvesting that may vary with area and degree of protection [81]. However, there is uncertainty on the contribution of illegal harvesting activities on the trophy size of these herbivores as in some instances some poachers tend to select horn size for their kills in the same manner as regulated trophy hunting [24]. Most illegal hunters (i.e., subsistence hunters) who target plains game, e.g., greater kudu and sable tend to kill indiscriminately and do not select for trophy size [26,27]. However, there has been worrying trends on illegal activities in Hwange region where illegal trophy hunting targeting African elephants has resorted to indiscriminate poisoning of large herds [23,28].
The recorded Cape buffalo trophy size in Matetsi Safari Area might not necessarily indicate trophy quality, but rather a possible limitation of the SCI scoring system [84]. Though a decline Trophy Hunting and Ecological Sustainability in Tropical Savanna Ecosystems in trophy size might suggest tendencies of unsustainable harvesting, a mature Cape buffalo bull has worn out horns that may produce low SCI tip to tip score [56,84]. The SCI scoring system in Cape buffalo mostly results in high scores for green bulls or soft bossed bulls (immature bulls) which are still in their breeding prime thus undermining the best practices in Cape buffalo hunting where only mature bulls, past their breeding prime and has broomed horns [84,85]. This maybe however different in the case of sable as there were declines in sable trophy size but with constant age at harvest in this study as also reported by Crosmary, Loveridge [55]. These variations may be due to habitat quality as a function of environmental heterogeneity that may influence resource allocation towards horn development and body growth over time [86][87][88]. Fluctuations in ecosystem productivity and habitat quality may result in cyclical trends in horn growth patterns observed in these species in Matetsi Safari Area as has been observed elsewhere [77].

Temporal and Spatial Harvesting Patterns
The basis for an increase in sable quota size in Matetsi Safari Area for the period 2004-2015 is problematic given reports on a possible decline of sable in its usual range within the Hwange Conservation Area [50]. Our findings cast doubt on the sustainability of how the quota setting processes in this area as there are indications that the quota allocations are not based on real scientific data. There seems to be over-reliance on questionable and subjective personal opinions in the quota setting process which in actual sense is supposed to be based on scientific evidence and ecological principles [46,47,89]. The trends observed in this study seem to reflect on the persistent use of the 'fixed quota' approach that tends to encourage harvesting of young or prime breeding individuals as an attempt by concessionaires to utilize the entire fixed portion of the quota regardless of its sustainability [46]. We argue that viability of trophy hunting in this area over time may be compromised unless solemn trophy hunting policy changes are adopted and implemented. Failure to utilize the allocated quota may reflect on (1) loss of hunting destination competiveness due to a waning preference by trophy hunting clients to patronise a hunting destination or species to hunt, (2) possible decline in the abundance of suitable trophy individuals thus affecting the hunting success of hunted species over time. It is argued that the viability of trophy hunting in this area over time may be compromised unless a review on the current trophy hunting policy is done to ensure that a dynamic framework is adopted and implemented commensurate with the global trends in modern day conservation. In this study, we report a spatial variation in trophy size attributes (i.e., age at harvest and trophy size) where Unit 6 had higher values for each of the attributes compared to Unit 1-5 except for Cape buffalo. Similar spatial variation in trophy size has been observed in the different provinces of South Africa [57]. However, as opposed to von Brandis and Reilly [56], our study was done in a more connected conservation area in the same region, where trophy hunting in Matetsi Area tends to utilize a shared, one source population given that there are no Trophy Hunting and Ecological Sustainability in Tropical Savanna Ecosystems fences in this area. We argue that Unit 6 may be having different attributes compared to other Units because could be benefiting from a source and sink dynamics associated with the movement of the selected species from Zambezi National Park and Unit 7 where there is no hunting. Moreover, Unit 6 could be benefiting from individuals migrating from Chobe National Park within the KAZA TFCA network. On the contrary, in areas where hunting has been persistent, animals have been observed to evolve avoidance mechanisms to evade disturbances and hunting [42,90]. Our observations in this study affirm the significance of sink and source dynamics of wildlife species between hunting and non-hunting areas within the KAZA TFCA network.
The historical and current trophy hunting activities in Matetsi might have shaped the antipredator strategies of these herbivores thereby avoiding the hunting areas in favour of the neighbouring National Parks within the KAZA TFCA where there is no hunting as was observed in Tanzania [91]. African elephant and Cape buffalo have a tendency of migrating within large landscapes in search of water and feed resources and this may result in the variation of trophy sizes observed in these units [92][93][94]. Within the KAZA TFCA landscape, we argue that though these harvesting rates may be considered low in relation to the population estimates of these species: they may not be sustainable from a trophy size perspective if age restrictions and trophy size limits were to be imposed.

Effect of Temporal Economic Status on Harvesting Patterns
In this study, there was no difference in the quota size allocation of Cape buffalo, African elephant, greater kudu and sable. Our results show that economic decline [95], may not affect the size of quota allocation levels in some hunting areas. However, it was evident that the economic decline also seriously incapacitated the Zimbabwe Parks and Wildlife Management Authority to do periodic surveys and monitoring programs which are critical in the quota setting process. Critical scientific data mostly obtained from aerial surveys in extensive and large protected areas were conducted in 2001 [96] and then recently in 2014 (http://www.greatelephant census. com). The quota sizes allocated during this period may have been based on previous experiences and individual opinions and not based on scientific principles as the MSY approach [39,40].
Although Zimbabwe had an economic crisis, the Zimbabwe Parks and Wildlife Management Authority did not substantially increase the trophy quota size to increase revenue for the Parks and Wildlife Estate to cushion itself from the bad economy. Instead, the Zimbabwe Parks and Wildlife Management Authority adapted by re-assigning Unit 2 and 4 to fall under its hunting concessions in 2013 and 2012 respectively as was the case with Unit 5. Similar management interventions have been observed in Sengwa Wildlife Research Area, which also falls under the Zimbabwe Parks and Wildlife Management Authority, where hunting was also recently introduced. In an attempt to increase its revenue base, the Zimbabwe Parks and Wildlife Management Authority over and above its regulatory role as an Authority is also responsible for hunting in Unit 2, 4 and 5 of Matetsi Safari Area.
We hereby argue that to some extent, the Zimbabwe Parks and Wildlife Management Authority relied on trophy hunting as a possible source of income for its operations as has been argued elsewhere [5,13,62]. The Zimbabwe Parks and Wildlife Management Authority is confronted with the dual task of generating revenue and yet at the same time plays the regulatory role in trophy hunting and wildlife conservation issues in Zimbabwe. Self-regulating is always a problem as there is often questions on 'who will police the regulator' and may cause problems if the regulator looses focus and allows the economic benefits to take precedence over regulatory policy framework [97]. As Zimbabwe recovers from the economic doldrums, there is need to seriously consider re-looking at the model which is being used in Matetsi Safari Area Unit 2, 4, and 5. This would promote transparency and accountability in the sustainable use of wildlife resources through trophy hunting.
In this study, there was no difference in the offtake levels between 2004-2008 and 2009-2015 time periods confirming the suggestions by Leader-Williams and Hutton [15] that political instability or economic decline may not reduce the trophy hunters' patronage to a hunting destination [95,98]. However, with the restrictive policy on import bans of elephant trophy from Zimbabwe into USA [9] and other restrictions by some European countries [99,100], it is clear that there will be a change in the proportion of hunters patronizing Zimbabwe for big game trophy hunting as the African elephant is one of the most sought for trophy species by most hunters. How this ban on the import of elephant trophies from Zimbabwe into USA as well as moratoriums by airlines in transporting such trophies would affect the trophy hunting in Zimbabwe and other southern African countries still need to be ascertained.

Conclusions and Recommendations
We concluded that: (1) the effect of trophy hunting on size of horn or tusk size and age at harvest is species specific as it does not necessarily affect trophy size and the age at harvest of harvested herbivores, (2) quota size allocation may not reflect the trophy size and offtake levels over the time, and (3) political and economic performance may not necessarily affect the harvesting regime patterns though external influences though moratoriums may possibly reduce the offtake levels over time. Accordingly, trophy hunting may not necessarily lead to irreversible trophy size over time but requires systematic monitoring and sound management interventions for sustainability [101].
We recommend that conservationists and protected area managers may consider to: (1) emphasise the need for ecological principles in the quota setting process and in some cases reduce or temporarily stopping hunting (i.e., introducing fallow or resting hunting years on a rotational basis) of some species, (2) create temporal and spatial refuges to facilitate 'trophy hunting rest' for some species promote reproductive to the desirable phenotypic sizes such as trophy size [102], (3) introduce and firmly implement age based harvesting policies across all trophy hunted species instead of lions only, through enforcing penalties for harvesting below threshold age individuals [46,89], and (4) introduce a variable trophy fee pricing system based on trophy size where the fees are determined on the trophy size [71,101]. These measures could then be replicated at micro-scale and national level in other areas where trophy hunting is being practiced to ensure sustainability.

Acknowledgments
This study was funded by Chinhoyi University of Technology, Grant Number PG4122. We acknowledge the Director-General of the Zimbabwe Parks and Wildlife Management Authority for allowing us to conduct this research and for the permission to publish this work. We thank the management staff and rangers of Matetsi Safari Area for their support during the course of our study. We are indebted to Mellinda Rushinga, Admire Chanyandura, Kudzai Mpakari, Thelma Chademana and Vongai Masunga who helped in the capturing and compilation of trophy size and harvesting regime data. Our special thanks to Chido Meda and Merillette M. Chihota for helping with the proofs. We acknowledge the comments from reviewers that improved the earlier versions of this paper particularly Staurt Primm and the Academic Editor, Suzannah Rutherford.