Accurate wildlife re-identification is critical for a wide range of ecological studies, including density estimation via capture-recapture, demographic analyses, and behavioral research. We pr esent GIRAFFE (Generalized Image-based Re-identification using AI for Fauna Feature Extraction), a system for automated re-identification of giraffes with extensibility to other species. Our approach uses local feature matching to identify known individuals and partition unknown individuals for label annotation at scale. Further, we develop a user interface that enables both technical and non-technical users to curate large datasets and analyze repeat survey data. In contrast with existing methods that require manual labeling to facilitate individual re-identification, GIRAFFE automates key steps in the re-identification pipeline, reducing manual effort while maintaining accuracy and interpretability. Validated on real-world giraffe datasets, the system achieves over 0.9 accuracy across nine standard metrics, with some metrics achieving near perfect score. It runs 120 times faster than baseline methods and delivers a 132-fold improvement in cost-effectiveness. This supports endangered species tracking, improves analysis of population dynamics and movement patterns, and, ultimately, allows for the implementation of data-driven conservation strategies.

Phenotype-dependent demography may result from heterogeneous selection pressures, which can be altered by temperature anomalies. Understanding which phenotypes are best adapted to different temperatures could clarify how populations may adapt to climatic changes and guide effective conservation measures. Animal markings are heritable phenotypic traits that show variation within populations, which suggest they may be adaptive. Variation in marking shape and size may influence thermoregulation and lead to differing individual responses to temperature anomalies. Therefore, markings may affect survival in variable and changing environments and be important for adapting to climate change. Using resighting data from 810 wild giraffes over 8 years, we estimated viability selection on spot patterns and whether it was affected by temperature anomalies. Calves with smaller lobate and adult males with smaller lobate or larger polygonal spots survived better. Additionally, calves and adult males with larger (vs. smaller) spots that experienced anomalously low (vs. high) temperatures survived better. Spot patterns had a smaller effect on adult females, with individuals of all spot types having lower survival in anomalously high temperatures. Synthesis and applications. Spot patterns influenced giraffe survival and their effects were altered by temperature anomalies. In calves, spot size may help with thermoregulation while spot shape may conceal them from predators. In adults, sex-specific selection pressures suggest sex differences in heat tolerance and trade-off with different functions. Conservation management maintaining variation in spot patterns by facilitating genetic exchange (e.g. through habitat connectivity) may help giraffes to adapt to climate change. Markings may have fitness consequences in other mammalian species and be important for population adaptation.

Responses of natural populations to climate change are driven by how multiple climatic and biotic factors affect survival and reproduction, and ultimately shape population dynamics. Yet, despite substantial progress in synthesizing the sensitivity of populations to climatic variation, comparative studies still overlook such complex interactions among drivers that generate variation in population-level metrics. Here, we use a common framework to synthesize how the joint effects of climate and biotic drivers on different vital rates impact population change, using unique long-term data from 41 species, ranging from trees to primates. We show that simultaneous effects of multiple climatic drivers exacerbate population responses to climate change, especially for fast-lived species. However, accounting for density feedbacks under climate variation buffers the effects of climate change on population dynamics. In all species considered in our analyses, such interactions between climate and density had starkly different effects depending on the age, size, or life-cycle stage of individuals, regardless of the life history of species. Our work provides the first general framework to assess how covarying effects of climate and density across a wide range of population models can impact populations of plants and animals under climate change.

Survival, reproduction, and movement are the key demographic parameters that drive population dynamics. Factors affecting these demographic parameters in large, long‐lived, extinction‐threatened mammals are diverse and may differentially affect subpopulations in disparate parts of an ecosystem. We conducted annual photographic surveys to uniquely identify 1,520 giraffes at 4 subpopulations around the Serengeti Ecosystem in Tanzania to estimate demographic parameters of age‐ and sex‐specific survival probabilities, reproduction, population densities, group sizes, and long‐distance movements. In the Seronera (central) subpopulation, we combined 15 years of data from 3 independent survey schemes, developed a Bayesian hidden Markov model to estimate demographic parameters, and conducted a retrospective population analysis to elucidate the demographic drivers of temporal changes in population growth rate. We collected data over 4–5 years for 3 other subpopulations, and used frequentist methods to estimate demographic parameters. We compared our results with historical estimates from the 1970s and 2000s to examine long‐term population trends and demographic drivers. We found significant differences in adult and subadult survival probabilities among subpopulations, with lower adult survival associated with declining subpopulations. Retrospective population analysis for the Seronera subpopulation reiterated that adult survival is a critical demographic driver of population dynamics for giraffes. The 2 subpopulations adjacent to the protected area boundary declined over 48 years, whereas the Seronera subpopulation stabilized since 2008. Only one individual moved between subpopulations, providing evidence for subpopulation insularity and potential genetic structuring of the overall population. These factors underscore the need for subpopulation‐specific conservation strategies aimed at raising adult survival within the western and northeastern parts of the Serengeti Ecosystem. Community‐based conservation efforts adjacent to protected areas have been effective in raising adult survival and density elsewhere. Our findings highlight the importance of understanding subpopulation dynamics and their demographic drivers for evidence‐based conservation and management to recover endangered giraffe populations.

An increase in the human population in recent years poses threats to the conservation of wildlife species. The expansion of human settlement and agricultural activities has led to the loss of wildlife corridors. Apart from the influence of humans on the gene flow of species between protected areas, natural features such as mountains, rivers, and hills can act as a barrier to gene flow. The African savanna elephant (Loxodonta africana) and the Masai giraffe (Giraffa tippelskirchi) are widely distributed in Tanzania. However, the Eastern Arc Mountains (EAM) and the Gregory Rift Valley (GRV) systems influence gene flow for some species. We conducted a study in southeastern Tanzania covering three major ecosystems: Ruaha-Rungwa, Katavi-Rukwa, and Selous-Mikumi to determine whether there is genetic differentiation between these ecosystems for giraffes and elephants. We analysed the mitochondrial DNA of 450 elephants and 100 giraffes. Our results show that (1) there is high genetic differentiation between populations found east and west of the EAM for both elephants and giraffes, (2) there is no female-mediated gene flow between these populations, (3) Populations found west of the EAM show high genetic connectivity suggesting historical gene flow between them, and (4) elephant populations from Ruaha share haplotypes with both Tarangire and Serengeti ecosystems suggesting historical connectivity between them. Our study reveals that the EAM plays a significant role in the gene flow of these species. However, the recent loss of miombo forests between Ruaha and Katavi, owing to anthropogenic activities, may reduce gene flow in the long run.

In East Africa, community-based conservation models (CBCMs) have been established to support the conservation of wildlife in fragmented landscapes like the Tarangire Ecosystem, Tanzania. To assess how different management approaches maintained large herbivore populations, we conducted line distance surveys and estimated seasonal densities of elephant, giraffe, zebra, and wildebeest in six management units, including three CBCMs, two national parks (positive controls), and one area with little conservation interventions (negative control). Using a Monte-Carlo approach to propagate uncertainties from the density estimates and trend analysis, we analyzed the resulting time series (2011-2019). Densities of the target species were consistently low in the site with little conservation interventions. In contrast, densities of zebra and wildebeest in CBCMs were similar to national parks, providing evidence that CBCMs contributed to the stabilization of these migratory populations in the central part of the ecosystem. CBCMs also supported giraffe and elephant densities similar to those found in national parks. In contrast, the functional connectivity of Lake Manyara National Park has not been augmented by CBCMs. Our analysis suggests that CBCMs can effectively conserve large herbivores, and that maintaining connectivity through CBCMs should be prioritized.

Many ungulate species in Africa range in habitats that vary in type and quality over space and time, but ongoing environmental change is substantially altering their habitats. Identifying key environmental variables that regulate ungulate population densities can guide management actions for effective conservation. We studied the local population density responses of a community of sympatric ungulate species in the Tarangire Ecosystem of northern Tanzania, to a suite of environmental factors that vary over space and time, to quantify population trends, determine the primary environmental correlates of densities, and identify covariation in densities among species. We estimated seasonal densities of five commonly detected species (impala, dik-dik, Grant’s gazelle, eland, and waterbuck) based on 7 years of distance-sampling data from 41 replicate surveys of 237 line transects. We systematically analyzed the effects of spatial, seasonal, and annual environmental covariates on variation in transect species-specific densities across space and time. Despite large fluctuations in climatic factors, we documented more spatial than temporal variation in four of the five species, suggesting that spatial heterogeneity may provide some buffer against temporal variation in the environment. Protection of sufficient habitats and water sources should allow ungulates to respond to a temporally changing world by moving across space. Further, among-species covariation patterns identified two potential ungulate guilds (impala—dik-dik—waterbuck; eland—grant’s gazelle) that should aid in developing efficient and coordinated management actions.

Giraffes exhibit a large sexual dimorphism in body size. Whether sexual dimorphisms also exist in body proportions of the axial and appendicular skeleton has been debated, particularly regarding the giraffe’s iconic long neck. We examined the anatomical proportions of the neck, forelegs, hindlegs, and body trunk of the Masai giraffe (G. tippelskirchi) in captive and wild populations. We found that female Masai giraffes have proportionally longer necks relative to their forelegs than males in contradiction to the original necks-for-sex hypothesis that proposed that the evolution of the giraffe’s long neck was driven by male-male competition. However, male neck width and apparent mass are proportionally larger than females’, supporting a modification of the necks-for-sex hypothesis. Moreover, male foreleg length is proportionally longer whereas female trunk length is proportionally longer. These sexual dimorphisms were found in both captive and wild Masai giraffes. We speculate that the initial evolution of the giraffe’s long neck and legs was driven by interspecific competition and the maternal nutritional demands of gestation and lactation through natural selection to gain a competitive advantage in browsing, and then later the neck mass was further increased as a consequence of male-male competition and sexual selection. Differences in the proportions of major body components define sex phenotypes, but several giraffes display opposite-sex phenotypes with a significantly higher level of discordancy seen in captive males. We speculate that body proportion sexual dimorphisms are maintained in the wild by natural and/or sexual selection, but in captivity selection is relaxed resulting in a higher occurrence of discordances in sexual phenotypes.

Abstract Giraffes exhibit a large sexual dimorphism in stature and body mass. Whether sexual dimorphisms also exist in relative body proportions of the axial and appendicular skeleton has been debated, particularly regarding the evolution of the giraffe’s iconic long neck. We measured and analyzed the relative anatomical proportions of the neck, legs, and body trunk of the Masai giraffe (G.c. tippelskirchi) in captive and wild populations. We found that female Masai giraffes have proportionally longer necks relative to their forelegs than males. Moreover, the female body trunk is proportionally longer whereas male foreleg length and neck width are proportionally greater. The sexual dimorphisms in body proportions were found in both captive and wild Masai giraffes suggesting that these differences are genetically determined. We speculate that the proportionally longer female neck is to compensate for females’ overall shorter stature to expand access to forage and their longer trunk is to accommodate fetal growth. Males’ longer forelegs, which contribute to the overall anterior body stature, likely provides some advantage in physical intrasexual competitions. Differences in the proportions of major body components define sex phenotypes, but several male and female giraffes display opposite-sex phenotypes with a significantly higher level of discordancy seen in captive males. We speculate that body proportion sexual dimorphisms are maintained in the wild by natural and/or sexual selection, but in captivity selection is relaxed because of human-altered mating and feeding behavior resulting in a higher proportion of sexual dimorphism discordances.

With the rapid pace of global warming, there is an urgent need to understand survival responses to climate, particularly for large mammals that are already experiencing population declines associated with anthropogenic pressures such as poaching and habitat loss. We tested hypotheses about the interactive effects of local climatic anomalies (variations around a long-term mean) and proximity to edge of protected area boundaries on seasonal adult and juvenile survival in a population of 2,385 individually identified giraffes monitored over 8 years in the Tarangire Ecosystem of northern Tanzania. Temperature anomalies were positively correlated with seasonal survival of adult giraffes, suggesting these megaherbivores are adapted to hot conditions. Higher seasonal rainfall anomalies were negatively correlated with both juvenile and adult survival, and greater vegetation greenness was associated with lower adult survival. During seasons of anomalously high rainfall and vegetation greenness, higher parasite and disease abundance, poorer-quality nutrition in forage, and higher predation risk may all play a role in lowering giraffe survival. Furthermore, climate-associated reduction in survival was most pronounced during the short rainy season for adult giraffes living closer to the edge of protected areas, indicating that the influence of climate anomalies may be exacerbated by anthropogenic edge effects such as poaching or livestock keeping. Precipitation in East Africa is projected to increase substantially, with a greater proportion of rain falling during heavy events in the short rainy season, which may threaten persistence of giraffes in one of Earth's most important landscapes for large mammals.

Megaherbivores play "outsized" roles in ecosystem functioning but are vulnerable to human impacts such as overhunting, land-use changes, and climate extremes. However, such impacts-and combinations of these impacts-on population dynamics are rarely examined using empirical data. To guide effective conservation actions under increasing global-change pressures, we developed a socially structured individual-based model (IBM) using long-term demographic data from female giraffes (Giraffa camelopardalis) in a human-influenced landscape in northern Tanzania, the Tarangire Ecosystem. This unfenced system includes savanna habitats with a wide gradient of anthropogenic pressures, from national parks, a wildlife ranch and community conservation areas, to unprotected village lands. We then simulated and projected over 50 years how realistic environmental and land-use management changes might affect this metapopulation of female giraffes. Scenarios included: (1) anthropogenic land-use changes including roads and agricultural/urban expansion; (2) reduction or improvement in wildlife law enforcement measures; (3) changes in populations of natural predators and migratory alternative prey; and (4) increases in rainfall as predicted for East Africa. The factor causing the greatest risk of rapid declines in female giraffe abundance in our simulations was a reduction in law enforcement leading to more poaching. Other threats decreased abundances of giraffes, but improving law enforcement in both of the study area's protected areas mitigated these impacts: a 0.01 increase in giraffe survival probability from improved law enforcement mitigated a 25% rise in heavy rainfall events by increasing abundance 19%, and mitigated the expansion of towns and blockage of dispersal movements by increasing abundance 22%. Our IBM enabled us to further quantify fine-scale abundance changes among female giraffe social communities, revealing potential source-sink interactions within the metapopulation. This flexible methodology can be adapted to test additional ecological questions in this landscape, or to model populations of giraffes or other species in different ecosystems.

Biofluorescence of mammalian pelage may serve to hide prey from predators sensitive to ultraviolet radiation, among other potential functions. To date biofluorescence has been documented in nocturnal-crepuscular and fossorial mammals that are active under low-light conditions. Giraffes are primarily diurnal, but biofluorescent pelage might offer camouflage from their nocturnal felid predators. Using a full-spectrum camera we qualitatively analyzed UV reflectance and absorption in giraffe pelage from a museum specimen. We found no trace of UV biofluorescence in the giraffe pelage, suggesting that this trait may not be ecologically or biologically relevant in giraffes. The function of biofluorescence in mammals remains elusive, but our study contributes to the growing body of data about biofluorescence, or its lack thereof, in diurnal versus nocturnal-crepuscular or fossorial mammal species.

The Masai giraffe has experienced a population decline from 70,000 to 35,000 in the past three decades and was declared an endangered subspecies by the IUCN in 2019. The remaining number of Masai giraffe are geographically separated by the steep cliffs of the Gregory Rift escarpments (GRE) in Tanzania and Kenya dividing them into two populations, one west and one east of the GRE. The cliffs of the GRE are formidable barriers to east-west dispersal and gene flow and the few remaining natural corridors through the GRE are occupied by human settlements. To assess the impact of the GRE on Masai giraffe gene flow, we examined whole genome sequences of nuclear and mitochondrial DNA (mtDNA) variation in populations located east (Tarangire ecosystem) and west (Serengeti ecosystem) of the GRE in northern Tanzania. Evidence from mtDNA variation, which measures female-mediated gene flow, suggests that females have not migrated across the GRE between populations in the Serengeti and Tarangire ecosystems in the past similar to 289,000 years. The analysis of nuclear DNA variation compared to mtDNA DNA variation suggests that male-mediated gene flow across the GRE has occurred more recently but stopped a few thousand years ago. Our findings show that Masai giraffes are split into two populations and fulfill the criteria for designation as distinct evolutionary significant units (ESUs), which we denote as western Masai giraffe and eastern Masai giraffe. While establishing giraffe dispersal corridors across the GRE is impractical, conservation efforts should be focused on maintaining connectivity among populations within each of these two populations. The importance of these efforts is heightened by our finding that the inbreeding coefficients are high in some of these Masai giraffe populations, which could result in inbreeding depression in the small and fragmented populations.

A population of Masai giraffes (Giraffa camelopardalis tippelskirchi) occurs in Arusha National Park (ANP), which is not part of the regular Tanzanian national wildlife monitoring scheme. Urban development of Arusha city and agricultural expansion have contributed to the increasing isolation of ANP from other protected areas in northern Tanzania. The only published data on the Masai giraffe population of ANP were individual-based data collected in 1979 and 1980. Here, we used individual-based data from 2021 to 2022 to provide an update on the current population size, population sex and age structure, movements and genetic connectivity of giraffes in ANP. We documented a 49% population decline and changes in the age distribution, adult sex ratio, reproductive rate and movement patterns relative to the previous study. Mitochondrial DNA analysis revealed genetic connectivity between ANP and other populations east of the Gregory Rift Escarpments in northern Tanzania and south-eastern Kenya, evidence that Masai giraffe once moved widely across the landscape.

Individual-based studies where animals are monitored through space and time enable explorations of ecology, demography, evolutionary biology, movements, and behavior. Here, we review 70 years of research on an endangered African herbivore, the giraffe, based on individual spot pattern recognition, and profile an example of a long-term photographic mark–recapture study of Masai giraffes in Tanzania. We illustrate how individual-based data can be used to examine the fitness consequences (variation in survival and reproduction) of extrinsic environmental factors or intrinsic traits in an evolutionary ecology framework. These data also allow the study of social structure, space use, life histories, and health. The giraffe offers an excellent opportunity to study dynamics of an ungulate species with a highly fission–fusion social system using spot pattern recognition.

For millennia, people have lived alongside wildlife in the semi-arid savanna of the Tarangire Ecosystem (TE), northern Tanzania. The TE preserves one of the last long-distance wildlife migrations in Africa as well as a large and diverse human population. Initial wildlife conservation approaches, settlement politics, and changes in human livelihoods have created a fragmented coupled social-ecological system that currently faces serious challenges for both people and wildlife. In this introduction to the book "Tarangire: Human-Wildlife Coexistence in a Fragmented Ecosystem" we outline the environmental and climatic settings as well as the social, economic, and political structures and histories of the ecosystem. The combination of heterogeneous geology, variable rainfall, a historical focus on conserving dry-season ranges of wildlife, and an expanding human population brings people and wildlife in contact, often with negative consequences for humans and wildlife. From an anthropocentric perspective, large carnivores and elephants are perceived as particularly problematic. In this book, we adopt a social-ecological approach and present different perspectives on wildlife conservation in the TE as frameworks for integrated and effective solutions. The first section of the book addresses the human dimension in human-wildlife interactions, whereas the second section employs a more ecocentric perspective and summarizes the status and ecologies of key large-mammal populations in the TE. The third section addresses human-wildlife interactions explicitly with an eye towards solutions.

Management of rangelands requires knowledge of forage species that are preferred or avoided by wildlife and livestock. A recent expansion of woody vegetation into previously open grasslands in African savanna ecosystems negatively impacts many mammalian grazers. Nevertheless, the ecological role of bush encroacher plant species as food may present a benefit for browsing species. We quantified diet selection by Masai giraffes (Giraffa camelopardalis tippelskirchi) through foraging observations and vegetation sampling in the Tarangire Ecosystem of Tanzania, which includes large areas of a native shrub that livestock managers have classified as an encroacher species (Dichrostachys cinerea). We compared woody plant species used by giraffes for foraging with availability at two different spatial scales during the wet and dry seasons. Giraffes selected some woody plants such as Vachellia species while significantly avoiding others, both at the local and landscape scales. Giraffes preferred foraging on D. cinerea at both spatial scales and in both the wet and dry seasons. Management that has focused on benefiting grazing livestock by removal of encroaching species (e.g., D. cinerea) may have unintended consequences for wildlife, especially for browsing species like giraffes that feed extensively on the expanding bush species.

Scientists are fallable and biased, but accuracy can be assessed through empirical analysis of published work that quantifies in-text citation (or quotation) errors. In scientific conflicts, it can be difficult for outsiders to know whose evidence or interpretation to trust. In-text citation error rate can assist decision- and policy-making bodies, as well as the courts when conflicts reach the judicial branch of government, by quantifying absolute and relative accuracy of scientists presenting scientific evidence. I propose the use of in-text citation error rates as a scientometric tool to quantify the accuracy of an author’s work. In-text citation error rates in excess of an established overall mean (e.g., 11% for minor errors and 7% for major errors in ecology), or differences in in-text citation error rates between opposing groups of scientists could be used to reveal excessive inaccuracies in an author or group. The spotted owl ( Strix occidentalis ) has been at the center of a multi-decadal conflict caused by competition among people over forest resources, with scientific experts representing opposing stakeholders often presenting conflicting evidence. I applied the in-text citation error rate tool to important papers in the spotted owl and forest fire debate and found evidence of greater error rates in works on one side of this debate. In-text citation error rate can be an effective tool for quantifying accuracy among scientists.

Savanna ecosystems support the highest diversities of hoofed mammal (ungulate) species in the world. Ungulates provide critical ecosystem services such as nutrient cycling and redistribution and play a key role in the food web, yet many species of ungulates are in decline due to anthropogenic activities. The fragmented Tarangire Ecosystem supports at least 25 wild ungulate species, yet few studies have been conducted on population status and habitat use in this region compared to the better-known Serengeti Ecosystem. In this chapter we review and discuss historical and current research on population trends of eight commonly detected species of ungulates in the Tarangire Ecosystem, and provide recommendations for long-term conservation of these culturally, economically, and ecologically important taxa.

Habitat selection is a dynamic biological process where species respond to spatiotemporal variation in resource availability. The resulting distribution patterns can be detected as presence–absence or heterogeneity in abundance and indicate habitat preferences based on environmental correlations at multiple scales. Variation in habitat selection by ungulates is constrained by trade-offs in top-down and bottom-up trophic processes arising from differences in forage requirements, water dependency, anthropogenic effects, and predation avoidance, and mediated by physiological (feeding guild) and morphological (body size) factors. We conducted distance sampling over 7 years in the Tarangire Ecosystem (TE) of northern Tanzania for six resident ungulate species: Kirk’s dik-dik (Madoqua kirkii), Grant’s gazelle (Nanger granti), Thomson’s gazelle (Eudorcas thomsonii), Masai giraffe (Giraffa camelopardalis tippelskirchi), impala (Aepyceros melampus), and common waterbuck (Kobus ellipsiprymnus), and tested hypotheses related to effects of top-down and bottom-up processes on ungulate presence and abundance. We modeled ecological correlates against two distributional responses to understand which environmental factors constrained these ungulate species at different scales; (i) presence–absence observations modeled in a logistic regression to assess habitat selection at an ecosystem scale; (ii) local abundances from presence-only observations modeled using a negative binomial distribution for finer-scale selection. Browser and grazer species in the TE selected suitable habitat proximal to rivers and avoided the Combretum–Azanza woody plant assemblage. Browsers and grazers also showed strong preference for habitat with more dense cover of preferred forage species, and abundance was influenced by the presence of specific forage species with significant seasonal variation. Mixed feeders were more heterogeneous in habitat suitability implying that broader diets allow avoidance of areas with high human activity. Small-bodied and dehydration-sensitive species selected areas near rivers and seasonal tributaries. Seasonal habitat selection was more pronounced among mixed feeders. Conservation strategies based on spatially and seasonally explicit resource selection studies such as ours can minimize impacts to biodiversity by protecting vital resources to ungulates through all seasons of the year.