«Item type text; Dissertation-Reproduction (electronic) Authors Munro, Natalie Dawn Publisher The University of Arizona. Rights Copyright © is held ...»
1998). The Ramat Yissakhar population provides an excellent example of high population growth. It occupies an agricultural landscape with an unnaturally rich distribution of food and water available year round due to agricultural activity.
Permanent food and water sources have increased both the survivorship and fertility of the Ramat Yissakhar gazelle population (Ayal and Baharav 1983; Baharav 1988), and thus its reproductive parameters represent the upper range of gazelle productivity. In contrast, the Ramat Qedesh population lives under more seasonal conditions and has a significantly lower rate of annual increase, if it grows at all. It is representative of a low growth situation, and thus well suited for application to the LGM. The following presents the parameters chosen for the gazelle simulations based on information available
Age at First Reproduction in Gazelles Female gazelles reach reproductive maturity during either their first or second year of life (between 6 and 18 months of age; Ayal and Baharav 1983; Baharav 1983a, 1983b; Shy et al. 1998). Gestation lasts for 6 months, and does bear their first fawn by the age of one or two years. The Ramat Yissakhar females become reproductively active at 6 months (Ayal and Baharav 1983; Baharav 1983b) and produce their first fawn at 12 months of age, while those from Ramat Qedesh often do not become pregnant until they reach 18 months of age. The discrepancy between the two populations is caused largely by differential access to permanent water sources. Other populations also report pregnancy in 6 month old does, though a relatively high percentage of these pregnancies were unsuccessful (e.g., 30%; Shy et al. 1998). Variation in the reproductive success of one year old gazelles is incorporated within the number of offspring variable, and an age of one year is adopted as the age of first reproduction in both the LGM and HGM Number of Gazelle Fawns Per Female Per Year Gazelle pregnancies most often lead to the birth of a single fawn (Ayal and Baharav 1983; Baharav 1983b). Twin births have never been observed in mountain gazelles, though they make up between 2.5 and 8.2% of live births in G. Siibgiittorosa Zhaowen et al. 1998), a closely related species. The gazelle birthing season is limited by the availability of standing water (Baharav 1983b). In well-watered locations, breeding is known to occur throughout the year, but peaks during the wettest months (Nowak 1991).
In mountain gazelle parturition is most often limited to a single discrete period in the late
The second birth season most likely represents a last attempt at reproduction by does who did not conceive in the spring. If water is abundant, does may produce more than one fawn per year. Under unusually favorable conditions (i.e., access to abundant standing water), mountain gazelle abandon rigid birthing seasons and increase productivity.
High productivity has been observed in mountain gazelles inhabiting the wellwatered regions of Ramat Yissakhar where recruitment reaches as many as 1.4 fawns per adult female each year (Ayal and Baharav 1983). Conversely, in neighboring regions, annual production is often less than a single fawn per female, depending on the proportion of failed pregnancies and access to water during the critical season. The lowest recorded birth rates in the studies examined here were 0.42 fawns per female per year in populations living in highly seasonal environments, however these rates were recorded in years when populations were in decline and may not be typical (Marraha 1996; Baharav 1983b).
While access to water clearly influences gazelle productivity, predicting the availability of standing water in the past is problematic. Artificial landscapes created by intensive agriculture inflate gazelle productivity in some cases by providing year round access to food and water. The lowering of the water table in recent years may have also reduced the availability of water to levels below those typical of the past. It is thus difficult to assess whether the reproductive rates of either of the two well-studied gazelle populations from Israel provide accurate representations of populations in the past. The
potential gazelle productivity, circumventing the problem of trying to create precise reconstructions of the past.
In the simulation, the minimum and maximum number of fawns produced per female per year are set at 0.7 and 1 fawn per female per year for the LGM, and at 1 to 1.4 fawns per female per year for the HGM. The number of fawns selected for the LGM is higher than some of the figures available for recent gazelle populations, because these populations were in decline. Using them for the simulations, thus creates an unviable population. The higher rate of 0.7 to 1 fawn per female per year is thus chosen as the minimum number of fawns required to maintain a stable population in accordance with the other LGM parameters.
Age of Onset of Adult Mortality The age of onset of adult mortality was derived in consideration of three points.
First, gazelle reach close to full body size by one year of age, although their bones continue to fuse until approximately 18 months of age (Davis 1980a). Second, gazelle reach reproductive maturity between 6 and 18 months of age. Finally, based on years of field observations, Baharav (1983b) estimates that gazelles are subject to adult mortality by the time they reach 1 year of age. The age of onset of adult mortality is thus set at 1 year for both models.
Mortality Mortality data for both juvenile and adult gazelles are seldom found in studies of
19831, 1983b, see also Ayal and Baharav 1983) provide high quality mortality data on the mountain gazelles from Ramat Yissakhar and Ramat Qedesh.
Juvenile Mortality The juvenile mortality rates for gazelles living under well-watered conditions at Ramat Yissakhar is 0.32 and is rounded down to 0.30 for use in the HGM. The juvenile mortality rate was much higher for the gazelle population surviving on natural resources at Ramat Qedesh (0.47) and is also rounded down slightly to 0.45 for use in the LGM model.
Adult Mortality The adult mortality values from the Ramat Yissakhar and Ramat Qedesh populations are recorded as 0.20 and 0.25 per annum respectively. In this case, the more productive Ramat Yissakhar population has a higher rate of adult mortality than the Ramat Qedesh population. This is the natural outcome of Ramat Yisshakhar's low rate of juvenile mortality. Because the simulations aim to model the extremes of population growth the lower rate (0.20) is used in the HGM simulation, but must also be used in the LGM simulation, since the higher mortality rate (0.25) outweighs annual fertility, and the population will not be viable in the long run.
Maximum Life Span Gazelles exceeding the age of 12 years of age have rarely been reported in the wild. Jones (1982) reports a G. dorcas that survived to the age of 17 in captivity.
extraordinary circumstances. Thus an maximum age of 12 years will be used for both the HGM and LGM models.
Results of the Gazelle Simulations: Age Structures and Hunting Pressure The gazelle population parameters were used to create stable LGM and HGM populations. Hunting was applied to these populations in gradual increments and the proportion of juveniles was recorded after each population stabilized. Juveniles are defined as individuals 18 months of age or younger to correspond to the oldest age for ftision of long bones, which are used to age Natufian gazelles in this study (see Chapter 8). The proportion of juveniles in the HGM and LGM populations subjected to incremental intensities of hunting pressure are plotted in Figures 6.6 and 6.7.
Three important observations can be made from these graphs. First the proportion of juveniles in the population increases gradually with hunting intensity in both the HGM and LGM populations (see Chapter 8). Second, the steepness of the graph slopes, representing the proportion of juvenile remains in the population, is greater for the HGM than it is for the LGM, since the HGM population is subjected to higher culls (up to 15% in comparison to 6% for the LGM). Third, and most importantly the models provide empirical estimates of the potential effects of hunting pressure on the living structures of viable gazelle populations. When no hunting is added to the simulations, the proportion of juvenile gazelles in both the HGM and LGM populations is about 30% (28% for the LGM and 31% for the HGM). This is the similar to the average proportion of juveniles
Figure 6.7: Proportion of juvenile gazelles (18 months or less) in the LGM gazelle population when subjected to increasing increments of hunting pressure.
Implications of the Gazelle Simulations for Monitoring Human Hunting Intensity The preceding simulations have defined the range of potential sustainable impact human hunters can exert on the living structures of gazelle populations. Because the
teeth collected by prehistoric hunters, the simulated ratios of juveniles to adults can be used as proxy measures of past human hunting pressure. Other factors ~ such as seasonality and hunting strategies ~ also influence the age structures of prey death assemblages. Defining the range of potential influence of each factor provides, however, a starting point for separating their role in assemblage formation. The simulations show that hunting pressure alone may inflate the proportion of juveniles in a stable gazelle population from approximately 30% to as high as 50% or perhaps even 60% in the case of heavily hunted high growth populations. Hunting at higher rates will crash the population, since like tortoises, gazelle populations have low rates of turnover. Finally, defining the living structure of prey populations means that hunted assemblages with percentages of juveniles outside of this range were most likely created by selective human hunting for specific age groups. The results of the gazelle simulations will be applied to
Questions of human demography — such as population pressure and sedentism— are difficult to test, yet they are of central interest in current Natufian research. This chapter presents a new method (following Stiner et al. 2000) for discerning finer-grained information about the intensity of site occupation. Tracking the proportions of small prey (tortoise, partridges, and hares) provides considerable resolution for detecting relative differences in the intensity of site use fi-om one place to the next and across two Natufian phases. Moreover, the abundance of small game relative to ungulate remains reveals the degree of pressure exerted by human populations on their environment at a regional scale (see Chapter 1).
To facilitate comparison between sites and time periods, assemblages are first grouped into three broad prey categories; ungulates, carnivores, and small game.
Although carnivores and predatory birds may not be "prey" senso stricto they were caught and used by humans. The small game fraction includes a variety of reptiles, birds, and small mammal species weighing no more than two kilograms. Only prey types
representation, fragmentation, and other taphonomic criteria, are included in this comparison (see Chapter 4).
Next, the small game index is applied (see Chapter 1). The small game fraction of each Natufian assemblage is subdivided into small mammals, birds, and reptiles based on arguments presented earlier in Chapter 4. Microfauna, including most rodents, small passerine birds, and the majority of small reptiles and amphibians have been shown to be intrusive, and thus are not included in the analysis. Specimens classified into general taxonomic groups such as small, medium, and large mammal are also excluded, because they could belong to any of two or more prey groups.
To address variation in hunting intensity across time and space in the Natufian, broad taxonomic and small game comparisons are presented. A summary of the Paleolithic faunal sequence at Hayonim Cave and Meged Rockshelter (Stiner et al. 1999,
2000) places the Natufian fauna in chronological and ecological perspective. The subdivision of the Natufian fauna from Hayonim Cave into five phase, allows the study of changes in the intensity of site occupation within the Natufian at this site. Finally, comparison of an expanded sample of Natufian sites including Hayonim Terrace, el-Wad Cave, and Hilazon Tachtit, with the Natufian of Hayonim Cave enables examination of regional trends within the Natufian period. This kind of temporal and regional framework is the foundation for discussing human demography at the Pleistocene/Holocene boundary. It also allows for more precise reconstruction of the
humans and their resources may have contributed to the onset of an agricultural revolution.
The large Natufian faunal sample from Hayonim Cave (NISP = 19,000) is subdivided to address questions posed on three different time scales. First, the Natufian fauna are treated as a single assemblage in order to bring out differences between the Natufian and preceding cultural adaptations, and allow examination of long-term change at Hayonim Cave beginning in the Middle Paleolithic (Stiner et al. 1999, 2000).
Although the Natufian layer is broken into more temporally refined units in subsequent analyses, the major differences that distinguish this assemblage from earlier Paleolithic layers are sustained. More fine-grained variation within the Natufian period at Hayonim Cave is discernible, however, when the assemblage is divided into a series of five consecutive phases (as defined by Bar-Yosef and Belfer-Cohen n.d.; Belfer-Cohen 1988;
and see Chapter 3). To strengthen the reliability of the Natufian analysis, fauna from problematic contexts and taxa were removed from the database, reducing the NISP count to 15,000 (see Chapter 3 for discussion of potentially mixed contexts, and Appendix I).
Finally, for inter-site analysis of the Natufian, the five phases from Hayonim Cave are collapsed into Early and Late Natufian categories. The Early Natufian is comprised of Phases I-III at Hayonim Cave, and the Late Natufian is equivalent to Phases IV and V (cf Bar-Yosef and Belfer-Cohen n.d.; Belfer-Cohen 1988).