OUR RESEARCH

Research as an essential DDCR management tool

Current research projects

The research we conduct at DDCR enables us to make management decisions based on sound scientific principles. It also provides baseline data for future surveys and studies.

Monitoring Program for the Major Site Values of the DDCR
Quantitative analysis of the hyper-arid vegetation changes under grazing in Dubai Desert Conservation Reserve
Quantitative analysis of the hyper-arid vegetation changes under grazing in Dubai Desert Conservation Reserve
Aims

To implement a continuous practical and efficient vegetation monitoring program by applying analytical tools for effective data interpretations. As vegetation is the base of the food pyramid the expected data and information is indispensable for managing the ecology of the Dubai Desert Conservation Reserve.

Goals

To evaluate the current state of the DDCR vegetation, the study is focuses on the assessment of density, cover, and biodiversity of the flora in the DDCR and follows the continued monitoring and repeated vegetation studies of 2004 (El Alqamy, 2004) and 2009 (Khafaga, 2009). To present the results through the mapping of the vegetation and by defining the lineage between different floral communities, using multivariate analysis combined with GIS techniques.

Methodology
Using the plot sampling method at 302 randomly selected sites, at each site, a circular plot is adapted for sampling the vegetation with an area of 50m diameter and each plot in total is equivalent to 7850m2. (Khafaga, 2009)

Plant species in each given plot are tentatively recorded in the field and put in tabulated form, giving the authentication of their identification with the help of the floristic books.

For each species total number of individuals are counted and an average size (length x width x height) is calculated.

To achieve the goal, data interpretation will require the adoption of the following strategies:

- Floristic composition and phytosociology
- Density and Relative Density
- Abundance and Relative Abundance
- Frequency and Relative Frequency
- Cover and Relative Cover
- Important Value Index (Species Dominance).
- Diversity Indices
- Simpson Dominance Index
- Shannon Diversity Index
- Margalef Richness Index
- Spatiotemporal variation of plant communities
- The calculation of the values of the diversity indices of individual plots over different habitats will interpolate predicted variety in un-sampled areas and will produce an overall picture of how diversity is changing over the entire DDCR.
Expected outcomes
The anticipated results will encompass three main themes. These are:
1. Assessing species diversity and variability of the DDCR flora.
2. Comparison of floral communities in different habitats, specifically comparing the populations of Gravel Plains against those of Sand Dunes.
3. Examining the vegetation diversity through various monitoring events and with altered adverse pressure and climatic factors.
Changes in functional diversity under grazing pressure in hyperarid shrubland ecosystem
Changes in functional diversity under grazing pressure in hyperarid shrubland ecosystem
Aims

To investigate the term Functional diversity “a variety of life-history traits presented by an assemblage of organisms” with regards to the vegetation diversity of DDCR. It has been postulated to be critical for the maintenance of ecosystem processes and properties.

The conceptual linkages between ecosystem disturbance, species diversity and functional diversity are fundamental to resource management and conservation planning.

Goals

To study the changes in intensity and timing of grazing and its effect on the changes in taxonomical and functional composition and subsequent changes in ecosystem functioning. These changes will explain the great interest in monitoring the impact of grazing on plant communities for the purpose of preserving their biodiversity and economic or aesthetic values.

Methodology
- Divide the reserve into three sectors (North, Centre and South).
- In each area, randomly select 25 sites for each habitat types (Gravel Plains and Sand Dunes), 150 sites in total.
- Build a matrix of different vegetation Functional traits for various monitoring times (2009 & 2017).
- Choose 24 “Functional Traits” to study the effect of grazing on vegetation and measure the response diversity in DDCR ecosystem.
- All attributes values will be recorded from published sources and herbarium information, except for the “plant heights” trait which will be measured in the field or collected from previous measured field studies.
- Anova and Diversity indices will be applied.
Expected outcomes
- A set of complete functional traits of DDCR.
- The change of the vegetation functional traits between different grazing pressures.
- A prediction of which functional character are more resilient and less vulnerable to any disturbances.
- Assessment of the changes in the diversity responses along a grazing gradient (the spatial variation of disturbance intensity).
Monitoring of the Arabian Oryx (Oryx leucoryx) in the Dubai Desert Conservation Reserve
Monitoring of the Arabian Oryx (Oryx leucoryx) in the Dubai Desert Conservation Reserve
Aims

To gain an understanding of the ecology of the Arabian Oryx within the DDCR, in particular we will look to determine their home range and which habitats, within that home range, they prefer and finally which plant species they utilize within that habitat.

Goals
- Gain a better understanding on the movements of Arabian Oryx through the use of GPS collars and so collect data on their home range and habitat selection.
- Gather data on activity patterns (i.e. time spent grazing, resting or moving) of the Arabian oryx in relation to factors such as habitat, temperature, sex and life phase.
- Gain an understanding of the dependence of the Oryx herd to the provided feed and how this can affect there behavioural and distributional patterns.
- Study the grazing habitats of the Arabian Oryx to establish grazing preferences in relation to plant species and growth form.
Methodology
- Eight Arabian Oryx, representing approximately 2% of the reserve population and consisting of four males and four females, will be selected from eight different herds geographically spread across the DDCR.
- The Oryx will be chemically immobilized and fitted with a GSM-GPS collar which will record the GPS location of the Oryx every two hours in addition an activity sensor records some behaviour such as grazing or resting. Collars should be active for three years before dropping off.
- Data from the collars will be transmitted, via GSM signal, to a base station collation and analysis.
- Current GPS locations and activity data (i.e. grazing periods), as well as VHF signals will be used to track the Oryx to ascertain their preferred fodder species.
Expected Output

Data from the collars will be analysed to establish home range, habitat preference and activity patterns in relation to factors such as sex, life phase and temperature. Furthermore by actively tracking Oryx in relation to grazing activity we will gain an understanding of habitat use and in particular plant forage preferences. By understanding the movement and habitat use of the Arabian Oryx, our largest herbivore, we will gain a better understanding of the herbivore carrying capacity of the DDCR and thereby make the correct management decision to prevent overgrazing and habitat degradation.
Monitoring of arid rangeland ecology using Unmanned Aerial Vehicles (UAVs) Desert Drones over Dubai
Monitoring of arid rangeland ecology using Unmanned Aerial Vehicles (UAVs) Desert Drones over Dubai
Aims

To establish a base in this emerging field within DDCR for the civilian use of aerial photography from Unmanned Aerial Vehicles (UAVs) and demonstrating its values for arid rangelands conservation.

Goals
Collecting accurate, inexpensive data to:
- Monitor grazing pressure on rangelands, and estimate botanic health of an arid ecosystem.
- Estimate and monitor the population of protected and endangered species
  
  Arabian Oryx, gazelle species, foxes, Gordon’s wild cats, Houbara bustard, birds of prey and Spiny-tailed lizards.
  
  Feral and invasive species, e.g. Prosopis juliflora.
- Evaluating the extent of damage caused by off-road recreational driving.
- Increasing the accuracy of ecological maps and consequently habitat classifications.
Methodology
After having the full set (Drone + Software) flights will be strictly limited to the 225km2 DDCR to avoid legal and privacy issues.
- Calibrating the airborne equipment over different altitudes, time of day and meteorological parameters to assess its abilities and limitations.
- Routine monitoring at selected elevations, to obtain spatio-temporal data of plant species and dune structure and possibly of some animals.
Each designated flight pattern will be recorded once per week for a year.
Expected Output

Temporal / Spatial baseline data of vegetation and dune change, related to topographic data; if the opportunity arises to study ephemeral species, then spatial distribution and growth of ephemerals over their lifespan.
Camera Trap Survey of the Dubai Desert Conservation Reserve
Camera Trap Survey of the Dubai Desert Conservation Reserve
Aims

As many species within desert environments are both nocturnal and elusive, it is difficult to gather reliable information on their populations. A camera trap is a camera that triggers when an animal passes in front of an infrared and motion detector. This has the advantage of detecting, with equal efficiency, both nocturnal and diurnal activities while having minimal environmental disturbance. The camera trap survey aims at recording the presence (or absence) of elusive and nocturnal species, in particular the smaller carnivores, within the DDCR.

Goals
- To collect photographic records from the deployed camera traps that can be used for analysis to establish species distribution and population size of the target species; Arabian Red Fox (Vulpes vulpes arabica), Sand Fox (Vulpes ruepelli), Gordon’s Wildcat (Felis silvestris gordoni) and Sand cat (Felis margarita).
- Determine which feral species are present and where are they concentrated.
- Identify and document all species recorded by the camera traps.
Methodology
- Nine camera traps, five Reconyx Rapid Fire and four Bushnell Trophy Cams, are deployed from the November 2010 until November 2011, into the five habitat types across the DDCR; Gravel Plains, Sandy Plains, Vegetated Sand Dunes, Shifting Sand Dunes and Rocky outcrop.
- Data, which includes photographs as well as date and time of the event, is collected from the camera traps every fortnightly. At the same time the locations are baited to attract the target species.
- The photographs are classified according content and species, if any animal was recorded. These classifications are then added to a spreadsheet with the other data, location, date and time.
Expected Output
- The establishment of either the presence or absence of the target species within the DDCR.
- Through statistical analysis of the data collected an estimation of the population size of the target species.
- On the basis of the data collected from the camera traps, maps of the record species distribution and density across the reserve.
- Habitat preference of all recorded species as well as information on patterns of activity of these species.
Assessment of Forage Productivity and Carrying Capacity
Assessment of Forage Productivity and Carrying Capacity
Aims

To investigate and recommend strategies, to manage and restore the rangelands and conserve their biodiversity in the Dubai Desert Conservation Reserve.

Goals
- Quantification of the forage productivity and evaluation of impact of important determining factors, such as soil and rainfall.
- Estimation of the rangelands carrying capacity, that is, the maximum possible stocking of livestock that a rangeland can support on a sustainable basis.
- Assessing the response of soil and vegetation to different grazing management practices for optimal rangeland use.
- Assessment of the possibility of restoring the degraded rangelands with indigenous forage plants and defining the most appropriate conditions for this process.
Methodology
- Site selection, sampling and environmental measurements.
- Estimation of forage productivity and carrying capacity.
- Building a model of the relationship between environmental factors and carrying capacity.
- Assessing the response of soil and vegetation to different grazing management practices for optimal rangeland use.
- Enclosures will be established.
- Number of plots will be located in each enclosure.
- Within each quadrate, the following plant community attributes will be recorded:
  
  A list for all the available species.
  
  Species density (number of individuals per unit area).
  
  Species frequency.
  
  Species cover (for perennials).
  
  Importance values (summation of relative frequency, relative density and relative cover) will be calculated for perennials.
  
  Species richness and diversity.
- Soil samples will be collected to measure its physical and chemical properties.
- Recovery of different species will be assessed in each enclosure.
Expected Output
- Development of management strategies and decision tools to proactively manage livestock grazing and reduce the drought impacts on plant community structure and function.
- Building a model that could predict the carrying capacity in any season or region.
- The rehabilitated sites would create more appropriate micro-habitats suitable for the reintroduction of some endangered or extinct in nature indigenous wildlife of the UAE.
Fixed Point Photography
Fixed Point Photography
Aims

To maintain a visual record of the different habitat types within the Dubai Desert Conservation Reserve.

Goals
- To have a series of standardized photographs, which can be compared on a season to season and year to year basis.
- Corresponding data on the fauna and flora of each site that can be used to gain a better understanding of the visual changes at each location.
Methodology
- Nineteen locations across four main habitat types of the DDCR were selected for a fixed-point photography project.
- A 360° view of each point will be photographed in the winter and summer of each year.
- At the same time, each point will have a rapid survey conducted on the area photographed, which will collect data on the diversity and abundance of fauna and flora.
Expected Output
- A visual record of the changes that occur to the different habitats within the DDCR, that can be explained by environmental factors (rainfall, temperature etc.) or management decisions (stocking rate, land use etc.). This could then be used as a tool for future decision making.
Macqueen’s Bustard Re-introduction
Macqueen’s Bustard Re-introduction
Aims

In a project done in conjunction with the National Avian Research Centre (NARC) and the office of HH Sheik Mohammed bin Rashid al Maktoum, Macqueen’s Bustard are to be re-introduced into Dubai Desert Conservation Reserve.

Goals
- To supplement the wild population, which migrate through the UAE, of this species that is classified, on the IUCN Red Data List, as vulnerable through rapid population decline.
Methodology
- Macqueen’s Bustards are captive bred in the National Avian Research Centre and reared to adulthood, before being transported to the DDCR.
- The bustards are release into the DDCR by one of two methods.
  
  Soft Release: The bustards are allowed to acclimatize to the area for one month in specially built cages before being released.
  
  Hard Release: The Bustards are released directly into the desert.
- Satellite tracked bustards are monitored via regular downloads and records kept of their movements.
Progress to Date
- The first release of 29 Macqueen’s bustards were released into the DDCR on the 25 February 2010. 15 bustards, including 3 with satellite tracking, were hard released directly into the desert in the South of the DDCR. While a further 14, including 3 with satellite tracking, were soft released near Tawi Ruwayyan after spending 1 month in cages erected on the release site.
- Since the first release an annual release of Macqueen’s Bustard has taken place except for in 2011.
- A total of 3101 individuals have been released with an approximate 1:1, male:female ratio up until the January 2020 release.
- 106 Macqueen’s Bustards have been tagged with GPS trackers and their movements studied by the National Avian Research Centre.
- Since 2010 breeding has been recorded in the DDCR through four nests with eggs being found and one female with chicks observed.
- Regular sightings of Macqueen’s bustards without satellite tracking are report by the DDCR staff, including a male doing a mating display.
Seasonal Dynamics of Herd Structure of Re-introduced Arabian Oryx in the Dubai Desert Conservation Reserve, United Arabian Emirates
Seasonal Dynamics of Herd Structure of Re-introduced Arabian Oryx in the Dubai Desert Conservation Reserve, United Arabian Emirates
Project Summary

Arabian Oryx (Oryx leucoryx, Pallas, 1777) is the largest species of antelope on the Arabian Peninsula. This magnificent animal was only saved from extinction through a captive breeding program and reintroduction to the wild. In Dubai Desert Conservation Reserve (UAE) antelope was successfully re-introduced in the 1999. Already in 2011 status of Arabian Oryx according to the IUCN Red List was changed from “Extinct in the wild” to “Vulnerable”.

Until 2019, the population of Arabian Oryx on the area of Conservation Reserve substantially increased (Lignereux, Alzahlawi, Al Kharusi and Pesci, 2018), however the knowledge on the dynamics in time and seasons is limited. Monitoring species abundance and distribution is a prerequisite when assessing species status and population viability. That is why the present research is needed. To describe the dynamics of Arabian Oryx population around feeding points, water holes, and on natural location (untouched); to understand the herd structure and influences driven by natural or artificial factors – all these are essential for successful development of management plan of protected area.

The main goal of the project is to identify the population dynamics over a selected time in relation to seasons, climatic conditions, locations and management measures (i.e. mobile feeding and water points). The period between 2009 and 2019 will be the main focus of the research. This timescale has close connection to present condition of the animal population and will help to prognoses the future population trend.

Research Objectives
- To identify density changes of re-introduced semi-wild population of Arabian oryx on the area of DDCR for the period 2009-2019.
- To treat available data from DDCR archive: images from camera traps, bio expedition reports; personal observation reports; climate stations.
- To identify the herd structure and activity patterns using the data from camera traps at different type of location within DDCR for the period 2009-2019.
- To interpret the Arabian oryx density and herd structure in relation to climate events using the data from climate stations throughout the investigated period.
- To analyze the continuos data from 2011 year with random camera trap distribution to determine the population dynamics. Compare population trends with discrete data from other years and locations.
Methodology
- In 2019, there are approx. 400 Arabian Oryx on the territory of DDCR. Eight of them have been equipped by GPS-collars, data from which are archived in the database of the DDCR. There is a collection of about 10,000 images from camera traps for the period from 2009 - 2019. Among these images, photos of Arabian Oryx will be identified and analyzed using custom computer based learning TensorFlow API and Python, Imagej, and CLIQS software packages. In the first stage of data processing, data from camera traps and climate station for the winter months 2019 will be mined and analyzed. Parameters such as number of individuals on the picture, their sex and age class (adult vs. calf), and activity will be recoded. The first analyze will show which approach and software design is the most suitable to reach the project goal.
- Personal observation at feeding and water stations will verify the assumptions about the herd structure and interpretations of activities recorded on pictures. To increase personal experience and gain real practical knowledge, skills and attitudes field work will be consisting of: bio expedition experience (non-invasive monitoring of animal), weekly observation and calculation of Arabian Oryx at DDCR, study visits to local centers, where Arabian Oryx breed in captivity. To understand the terrain characteristic in real time (and as result - cover more behavioral patterns within different climatic conditions) field work will be splitted into 2 periods: summer and winter. Personal present in the office of DDCR will be helpful as well during the computing analyze of data with supervision of DDCR experts.
- Data from climate stations will provide information about the regular climate pattern and unusual climatic events on the territory from 2009-2019. The animal density dynamics will be tested for the effects of climatic parameters using relevant statistical procedures. Finally, the results of animal density and spatial distribution of feeding and water points will be visualized using tools in ArcGIS Pro software (e.g. Space Time Cube).
- Understanding of seasonal dynamic of herd structure and its dependence on natural, artificial factors will provide DDCR team with essential information for further managerial decisions.
- The analyze of managerial decisions (dislocation of feeding points) in the period of 2009-2019 will help to create the map of already-used and non-used plots on the DDCR area. This map will give a base to conservation management plan for future decision.
Expected Outcomes
- Information about seasonal dynamic of herd structure for the period 2009-2019 will be analyzed and prognosis of future population dynamic will be formulated. Effects of climate conditions on animal dynamics throughout the time period (2009-2019) will be visualized via Space Time Cube in ArcMap.
- Data findings at completion will be presented to the research committee in the form of presentation.
- Research results, as well as general research project outline will be organized as visual content of digital poster. Poster can be published on official website, media channels of DDCR.
- The recommendation to Management Conservation Plan will be made, based on the information about seasonal dynamic of herd structure and its dependence on natural, artificial factors.
- Final results of research will be presented to the academic community of Czech University of Life Science (Prague) in the form of presentation.
The Genetic Study of the Arabian Oryx Population in Dubai, United Arab Emirates
The Genetic Study of the Arabian Oryx Population in Dubai, United Arab Emirates
Project Summary
Over the past 20 years the breeding of Arabian Oryx has been very successful, both in Dubai and around the world. Bring the species back from the state of Extinct in the wild and classified as Endangered by the IUCN Red List to having healthy re-introduced populations in a number of countries have led to a downgrade of its endangered status to Vulnerable. However it is important to now change the objective from reproduction to a sustainable population by ensuring we have the most genetically diverse and viable population possible to ensure their long term survival. This research aims to achieve the following;
- Ascertain the genetic diversity of the Oryx population found within the Dubai Emirate.
- Utilize the genetic data to develop a strategy to improve the genetic quality of Dubai’s Arabian Oryx population.
- Contribute to the regional efforts for the conservation of Arabian Oryx.
Research Objectives
- Collect blood, Tissue or Plucked hair samples from 10% of the Dubai population.
- Complete genetic analysis on all of these samples.
- Establish the level of genetic diversity within the Dubai population and measure it against the genetic diversity and maternal lines of genetically known populations.
- Design a genetic management program to enhance the genetic diversity of the Dubai population as well as contributing to other regional programs.
Methodology

There are approximately 2000 individual in the Arabian Oryx population of Dubai this comprises collections under intensive management as well as populations in extensive Protected areas. As stated the goal will be to collect either Tissue or blood samples from 10% of the population as per the University of Edinburgh, Arabian Oryx Genetic Management – Sampling Protocol (attached). Samples will be collected from as many different herds as possible within each collection. Tissue samples will be collected with a biopsy needle shot from a dart gun and blood samples will only be collected if the animals are being handled for other reasons.

Expected Outcomes
- A report will be prepared by University of Edinburgh, Conservation Genetics Lab. This report will detail the results of the DNA extraction and sequencing, providing us with a genetic profile of the Arabian population within Dubai.
- The data gathered from the genetic report will be utilized to develop a plan to maximize the genetic diversity of the Dubai oryx population. This will be completed by assessing the differences in genetic material between the different herds in Dubai as well as other populations with known genetics and developing a plan of exchange to maximize genetic diversity in the DDCR herd as well as the other herds in Dubai and the region.
Machine Learning for Bio-diversity Monitoring and Tracking at Scale
Machine Learning for Bio-diversity Monitoring and Tracking at Scale
Project Summary

The need to protect the environment has taken center stage in recent years as the damage to the Earth’s natural ecological balance has become increasingly tangible. Natural reserves are set to protect the biodiversity. With this good intent comes great challenges. Understanding the populations of animals and vegetation living inside the reserve is a key parameter to secure the equilibrium and the continuity of the ecosystem. We propose to investigate the problem of biodiversity monitoring and tracking using machine learning techniques for mining useful information from complex data, image/videos.

Deep learning methods have proven to be efficient in the image processing field. In fact, their introduction in this area has reshaped the way images are processed and understood. At the same time, this raised expectations with regards to performing and automating the processing and understanding of images, along with understanding their content. A large number of works have been performed and many methods have been proposed to tackle different aspects of the problem. The novelty in this project resides in two folds: it is the first time that the performances of existing algorithms will be evaluated in such large magnitude and diversity of experiments, which will support the investigators to propose new algorithms. The second novelty resides in the application of the approach in the area of biodiversity tracking and monitoring in the desert. In fact, the investigators are not aware of any existing efforts that suggest the application of image mining in the desert.

The investigators intend to work closely with the DDCR to help detect and monitor the different species, their evolution, movement, and gender. This will directly help the DDCR in better estimating the number of animals, predict their evolution, understand their moves, and understand the density of those animals. The project is expected to be beneficial for the wildlife and the biodiversity.

Research Objectives
- Collect a significant quantity of data to test a number of existing methods in real-life situations.
- Determine benchmarks to be used in the evaluation of future methods.
- Setup optimized techniques for scanning large desert areas for resources management.
- Propose methods for identifying animals, their family, gender, and movements.
- Create a set of methods and workflows to implement the findings of this project.
Methodology
This research will make use of modern image mining techniques to contribute to the area of environmental protection through automatic animals’ recognition and identification in wild areas. In this project, we will begin by creating a large scale and controlled experiments protocol to evaluate the top existing methods in research w.r.t. our context, i.e., animals identification in wild areas. We will then follow an experiment-based methodology in which real situations and real datasets (e.g., UAV captured images and videos) will be used for observation and learning models. These datasets will serve as a foundation for enabling the extraction of actionable knowledge. Once the experiments are performed; we will proceed to researching a method (or methods) that would integrate the knowledge and the learnings from the experimental phase.

At a high level, we will follow the steps shown in the next section.
- Data collection and preparation: As a backbone of our research and experiments, the data to analyze will be related to litter in wild areas, i.e., desert in our case. As UAVs will be largely used here, the configuration of the experiments must be taken care of to ensure an objective assessment and interpretable quality of the outcomes. The datasets have to be tagged and classified to make sure that we can build the expected models. Considering the nature of the environment, different attempts may be considered to ensure a high quality of data is gotten.
- Benchmarking of existing methods: Relying on the collected data, a large experimental setup will be put in place to evaluate the most recent and advanced techniques available in the literature to identify animals, as well as movements and related concepts.
- Model(s) development: The objective here is to elaborate new methods for supporting animals identification in wild areas. While developing the models, we should take into consideration that we are dealing with real-time and heterogeneous data taken under different constraints.
- Implementation and Evaluation: This step will basically run across the entire project. Its objective will be to define the quality of the proposed methods and approaches in terms of effectiveness and performance. We will perform this objective through formal methods and subjective evaluations.
Expected Outcomes
- The project will provide a set of data, described in detail and capturing different situations and constraints.
- This report will contain a rigorous description of a set of experimental protocols that will govern the test of different methods and techniques in the literature.
- This deliverable is the final outcome and expectation of the project. It is a software platform that is dedicated for end-users and which will allow them to analyze real-time streams coming from unmanned aerial vehicles. It will embed artificial intelligence techniques that will not only reduce, e.g., the noise in image streams but also the selection of the best configuration to use by the UAV depending on the external constraints. The software is expected at the end of the project in its final version.
- We will target at least 2 top conferences/journals publications. These publications will be expected during the second year of the project, mainly.
- This report is a classical final report. It will be provided at the end of the project and will describe the achieved objectives and outcomes of the project.
Large Scale Explorations of Real-Time Image Mining Captured by Unmanned Aerial Vehicles (UAVs): Application to Environment Protection
Large Scale Explorations of Real-Time Image Mining Captured by Unmanned Aerial Vehicles (UAVs): Application to Environment Protection
Project Summary

The need to protect the environment has taken center stage in recent years as the damage to the Earth’s natural ecological balance has become increasingly tangible. As humans, we consume a large amount of plastic as it has become a cornerstone of many manufacturing and packaging processes. In fact, it is common practice to dispose of unwanted waste improperly in the form of litter and illegally dumped garbage. Whether this occurs on the streets, where many people simply drop their unwanted packaging, or during time spent in extra-urban environments, it is clear that littering has become a significant problem in many societies and contexts.

Deep learning methods have proven to be efficient in the image processing field. In fact, their introduction in this area has reshaped the way images are processed and understood. At the same time, this raised expectations with regards to performing and automating the processing and understanding of images, along with understanding their content. A large number of works have been performed and many methods have been proposed to tackle different aspects of the problem. The novelty in this project resides in two folds: it is the first time that the performances of existing algorithms will be evaluated in such large magnitude and diversity of experiments, which will support the investigators to propose new algorithms. The second novelty resides in the application of approach in the area of liter identification in the desert. In fact, the investigators are not aware of any existing efforts that suggest the application of image mining in the desert.

The investigators intend to work closely with the DDCR to help detect and monitor litter in three key areas; along the boundary fence, camp areas and driving routes throughout the reserve. This will directly aid the DDCR in coordinating litter recovery as well issuing fines to parties who have violated the regulations regarding litter. In addition to a potential monetary gain and reduction in resources dispensed in patrolling the reserve, the wildlife will benefit from a reduced exposure to harmful objects which they may mistake for food.

Research Objectives
- Collect a significant quantity of data to test a number of existing methods in real-life situations.
- Determine the constraints that affect the quality of image understanding in real situations.
- Determine benchmarks to be used in the evaluation of future methods.
- Create a set of methods and workflows to implement the findings of this project.
Methodology
This research will make use of modern image mining techniques to contribute to the area of environmental protection through automatic litter recognition and identification in wild areas. In this project, we will begin by creating a large scale and controlled experiments protocol to evaluate the top existing methods in research w.r.t. our context, i.e., litter identification in wild areas. We will then follow an experiment-based methodology in which real situations and real datasets (e.g., UAV captured images and videos) will be used for observation and learning models. These datasets will serve as a foundation for enabling the extraction of actionable knowledge. Once the experiments are performed; we will proceed to researching a method (or methods) that would integrate the knowledge and the learnings from the experimental phase.

At a high level, we will follow the steps shown in the next section.
- Data collection and preparation: As a backbone of our research and experiments, the data to analyze will be related to litter in wild areas, i.e., desert in our case. As UAVs will be largely used here, the configuration of the experiments must be taken care of to ensure an objective assessment and interpretable quality of the outcomes. The datasets have to be tagged and classified to make sure that we can build the expected models.
- Benchmarking of existing methods: Relying on the collected data, a large experimental setup will be put in place to evaluate the most recent and advanced techniques available in the literature to identify litter.
- Model(s) development: The objective here is to elaborate new methods for supporting litter identification in wild areas. While developing the models, we should take into consideration that we are dealing with real-time and heterogeneous data taken under different constraints.
- Implementation and Evaluation: This step will basically run across the entire project. Its objective will be to define the quality of the proposed methods and approaches in terms of effectiveness and performance. We will perform this objective through formal methods and subjective evaluations.
Expected Outcomes
- The project will provide a set of data, described in detail and capturing different situations and constraints.
- This report will contain a rigorous description of a set of experimental protocols that will govern the test of different methods and techniques in the literature.
- This deliverable is the final outcome and expectation of the project. It is a software platform that is dedicated for end-users and which will allow them to analyze real-time streams coming from unmanned aerial vehicles. It will embed artificial intelligence techniques that will not only reduce, e.g., the noise in image streams but also the selection of the best configuration to use by the UAV depending on the external constraints. The software is expected at the end of the project in its final version.
- We will target at least 2 top conferences/journals publications. These publications will be expected during the second year of the project, mainly.
- This report is a classical final report. It will be provided at the end of the project and will describe the achieved objectives and outcomes of the project.
Pathogen Risk Assessment on Arabian felines in the United Arab Emirates and Oman
Pathogen Risk Assessment on Arabian felines in the United Arab Emirates and Oman
Project Summary

Arabian felines (Arabian leopards, cheetahs, wildcats, sand cats and caracals) are either regionally extinct, critically endangered or their populations are worryingly declining in the Arabian Peninsula. Thus, the International Union for Conservation of Nature (IUCN), the global authority on the status of the natural world and the measures needed to safeguard it, claimed the need to carry out further research on the factors that may threaten these five felids.

Felid conservation will increasingly need to include assessments of disease risks and strategies for disease management in order to be successful, since it is known that most wild-feline populations are at risk of exposure to new pathogens due to deterioration of many of their inhabited ecosystems, as well as the fact that historic geographic barriers have been bridged, which might imperil population health. At the same time, due to the gravity of these animals’ situation in the wild, they are held and bred in captivity at centres worldwide. Such captive breeding programmes are essential in maintaining genetic diversity and saving these species from extinction, since their reintroduction may be the only means of restoring their populations in the wild. Nevertheless, captive animals are more likely to contract diseases because animals live in close proximity to each other. Whereas in nature diseases that are caused by parasites comprise one of the major problems resulting in even morbidity and mortality in these animals, in captivity the effects range from sub-clinical to death.

In this framework, the aim of this project would be to determine the types, prevalence, and intensity of parasitic, bacterial and viral infections in the wild and captive Arabian felines located in the United Arab Emirates and Oman. Very little has so far been published on pathogenic diseases of these felines in the Arabian Peninsula either in the wild or in captivity, and a study of this scope in such field would be the first of its kind. The results obtained will constitute a baseline data and a vital reference for future research in such neglected area. Moreover, the knowledge obtained from this project will, inevitably, be used in future reintroduction and reinforcement projects of these species in both countries.

Research Objectives
- Which parasitic, bacterial and viral pathogen species infect/circulate within the Arabian felines populations in the UAE and Oman?
- When comparing wild and captive Arabian felines populations, what are the differences in pathogen prevalence and abundance?
- Does the prevalence and/or abundance of some of these pathogens comprise a threat to the Arabian felines’ lives?
Methodology

Please see Appendix I and II.

Expected Outcomes
- Launching collaboration between the Veterinary Medicine Department and Department of Biology at UAEU, and The Biology Department at SQU, as well as Governmental authorities and NGOs working in conservation of the Arabian felines in both countries.
- The expected results will be published in peer reviewed journals related to animal conservation and health.
- The data and conclusions will be presented in national and international conferences.
Statistical evaluation of Sorbonne University Atmospheric Forecasting System (SUAFS) at Dubai Desert Conservation Reserve (DDCR) (phase 1)
Statistical evaluation of Sorbonne University Atmospheric Forecasting System (SUAFS) at Dubai Desert Conservation Reserve (DDCR) (phase 1)
Project Summary
Sorbonne University Atmospheric Forecasting System (SUAFS) is an integrated combination of advanced models able to provide detailed weather desert forecasts on various temporal scales over the Arabian Peninsula and the greater UAE areas.. The following predictions currently available by SUAFS (http://forecast.psuad.ac.ae) are:
- Weather forecast for 72 hrs ahead (3 days).
- Desert dust forecast for 72 hrs ahead (3 days).
SUAFS is running operationally since July 2018 and an annual archive of weather and desert dust forecasts are almost available for evaluation.

This research proposal offers a unique opportunity to initiate the framework for an important collaboration with DDCR in terms of environmental warnings and monitoring. Thus, the main aim of the Phase-1 project is to validate SUAFS forecasts against the atmospheric measurements obtained from DDCR meteorological stations. The entire records and data of the main atmospheric parameters from the network of the meteorological stations of DDCR, as well as the neighbored available stations, will be utilized for comparison with SUAFS forecasts. The extracting statistical scores will be helpful to identify the forecast errors of the system and their propagation into the period of the simulation. The results of this research work will lead SUAFS forecasts to achieve greater accuracy and reliability especially over the area of interest.

These results will be the base of the Phase 2, who focus on the air quality monitoring and dust forecasting dedicating for the DDCR. All the results will be at DDCR and SUAD disposal and a report with all the data will be submitted. Phase-1 will also act as a preparatory phase for attracting additional funding to launch the demanding Phase-2.
Research Objectives
- Process and homogenization of the atmospheric measurements from the DDCR meteorological stations network.
- Statistical evaluation of SUAFS.
- Achievement of greater accuracy and reliability for the weather forecasting conditions in DDCR broader area.
Methodology

In Phase-1, the primitive records (bulk data) obtained from the meteorological stations will be processed in order to be implemented for the SUAFS assessment. Atmospheric forecasts of specific parameters such as near surface temperature, wind speed, humidity, mean sea level pressure and precipitation will be also processed for direct comparison against in-situ records of the stations.

The evaluation of the SUAFS weather forecasts will be made in the DDCR area where the available surface stations have been installed. Additionally, surface observations from neighborhood networks will be used to verify and compare categorical model forecasts of the 10-m wind field, 2-m air temperature and sea level pressure and the accumulated precipitation.

The evaluation methodology will be based on the point-to-point comparison between model-generated variables and observations. For the variables of wind speed, air temperature, humidity and sea level pressure the scores produced are the standard mean error (BIAS) and the root mean square error (RMSE). The verification scores used for the precipitation will be derived using the contingency table approach. This is a two-dimensional matrix where each element counts the number of occurrences in which the gauge measurements and the model forecasts exceeded or failed to reach a certain threshold for a given forecast period. Thus, the forecast skill can be measured by evaluating the bias score (BS) and the equitable threat score (ETS).

Expected Outcomes
- SUAFS statistical scores (BIAS, RMSE, ETS) in the forecast window of 06-72hours
- Final report or publication in scientific journal or in conference proceedings. Presentation or one-day workshop in DDCR or SUAD to present the results.
Analysis of trophic niche partitioning of desert rodents through diet reconstruction
Analysis of trophic niche partitioning of desert rodents through diet reconstruction
Project Summary

For this project we want to investigate the trophic niche partitioning of the desert rodents found in the Dubai Desert Conservation Reserve. Currently, the rodent populations and inter-species interactions in the DDCR are relatively unknown. We plan to use next generation sequencing (NGS) on fecal samples from the rodents species found in the DDCR to analyze their diet, gut microbiomes and parasite composition in order to evaluate the presence of trophic niche partitioning and to determine the functional role the gut microbe communities play in host evolution.

Furthermore, we want to evaluate the effects of anthropogenic features in the reserve (i.e. feeding stations and the Al Maha resort) on the rodents’ diets by looking for significant differences in our data from both far and near these features. We want to provide a deeper understanding of UAE desert ecological structures, their dynamics, and mechanisms. This is especially relevant for any future conservation or management action implemented on the area or on the rodents themselves.

Research Objectives
- Reconstruct the typical diet of the six rodent species using NGS from fecal samples and compare the results for the presence of trophic partitioning.
- Examine the effects of grazers’ feeding stations on the desert rodents.
- Investigate if there is a significant anthropogenic effect from the Al Maha resort on the rodents’ diets.
Methodology

Field study: we plan to deploy approximately 100 baited rodent traps (Sherman traps, life traps) in the DDCR area. The traps will be equally divided amongst the different habitats, as well as close and far from the feeding stations and close and far from the resort. We will conduct three field visits during three different seasons lasting between 2 and 3 days per visit. We expect to obtain 20 feces and tissue samples (obtained via ear punching; for DNA barcoding) per rodent species per season; with a minimum of 5 samples per species per season for sufficient statistical analysis. Additionally, we will collect the available plant species in the different rodent habitats (i.e. the food resource) from the DDCR for DNA barcoding and creation of a reference catalog.

From the samples we will extract the DNA and run 18s and 16s PCR-based amplification. We will do 18s and 16s gene metabarcoding using next generation sequencing. We will analyze the results for presence and absence of organisms and relative abundance in order to check for partitioning of the niche. Furthermore, we will look for significant differences in dietary choices of rodents influenced either by the resort or the feeding stations.

Most statistical analyses will be conducted using Permanova and partial Mantel tests using R software.

Expected Outcomes
- Using next generation sequencing we will obtain the range of the diet of the desert rodents and whether there is trophic niche partitioning.
- Analysis of effects of anthropogenic actions on the rodents.
Understanding Spatiotemporal Dynamics of a Desert Gecko Community. Insights From Diet, Microbiome, and Parasitic Composition
Understanding Spatiotemporal Dynamics of a Desert Gecko Community. Insights From Diet, Microbiome, and Parasitic Composition
Project Summary

Understanding trophic interactions and the ways in which they influence niche partitioning remain at the core of ecology, and may have strong implications for conservation biology. Unfortunately, the dynamics of trophic interactions within communities of desert species remain largely undescribed. An example reflecting this claim is a community of nocturnal geckos inhabiting sandy sheets of the United Arab Emirates, which consists of three species - Stenodactylus arabicus, Stenodactylus doriae and Bunopus tuberculatus. A single study investigating basic ecology of the desert lizards of Arabia characterized the aforementioned geckos as insectivores. However, the exact way in which the partitioning of the dietary niche occurs within this community remains unknown.

A similar lack of understanding occurs within a system of two species inhabiting gravel plains. Stenodactylus slevini and Stenodactylus leptocosymbotes are two closely related nocturnal sepcies characterized by comparable morphological traits. Interestingly, field surveys revealed lack of overlap in the occurrence ranges, indicating that these two gecko species may exclude each other. An explanation of this phenomenon may be a potential overlap in dietary prefrences, however, this claim has never been investigated.

Diet is also known to influence microbial and parasitic composition. The vast majority of microorganisms associated with vertebrates can be found in the digestive system. These microorganisms have been shown to influence the host's physiology, development, and immune system. Until today, it remains a major goal to determine the functional role these microbe communities play in host evolution.

Excessive grazing of captive camels has been recognized as the single greatest threat to the desert ecosystems of the UAE. However, the effect of this phenomenon on small vertebrate (i.e lizard) communities remains unknown. Even though in most cases a direct prey competition between these animals and camels does not occur, there may be a strong indirect effect exerted by the grazers (decrease in number of prey animals who are dependent on substantial vegetation cover, reduction of hiding places, increase in terrestrial temperatures, etc). Hence, it remains of uttermost importance to understand how the results of overgrazing affect the dynamics of small vertebrate communities in order to establish appropriate conservational measures.

The proposed study aims at identifying the dietary preferences, microbial and parasitic compositions of the gecko species across different seasons and locations with varying grazing pressure. For this, stool samples of S. arabicus, S. doriae and B. tuberculatus will be collected over three seasons (fall, winter, summer) and contents will be analyzed using next-generation sequencing. In addition, prey availability and selection will be analyzed using pitfall trapping of arthropods and selective sequencing for barcoding purposes. Sampling will be conducted in two distinct locations - the DDCR and the Arabian Nights Village Road (Abu Dhabi Emirate), which differ significantly in the extent of vegetation cover. The project will potentially be extended to include S. slevini and S. leptocosymbotes from the gravel plains to determine reasons for mutual exclusion.

The results from next-generation sequencing will be analyzed using Permanova and partial Mantel tests to detect significant differences in diet, microbial and parasite compositions between the study species.

Research Objectives
- To identify the dietary preferences of the three gecko species inhabiting the sandy sheets (B. tuberculatus, S. doriae, S. arabicus) across seasons.
- To determine the effect of dietary preference on niche partitioning of S. slevini and S. leptocosymbotes.
- To identify the effect of vegetation cover on the dietary preferences, microbiome and parasitic composition of the gecko species.
Methodology

Individuals from the chosen gecko species will be collected at night in order to obtain their fecal samples. The fresh stool will be transferred to sterile tubes and stored in liquid nitrogen immediately. The sampled geckos will be individually marked in order to avoid repetitions. Additionally, the sampling locations will be surveyed for insects, which will be captured, identified and euthanized in order to obtain 18s sequences.

Upon return to the laboratory, DNA from the feces will be extracted using Qiagen Stool Kit. Then, genetic markers 16s and 18s will be amplified using PCR-based protocol. The PRC products will be sequenced using next-generation sequencing. The sequences retrieved from the facility will be annotated, cleaned and compared with the ones available in the datasets. In such a way, species-specific information about dietary preferences, microbial and parasitic composition will be obtained and used to answer the questions at the community level.

Expected Outcomes
- We expect that closely related, syntopic, and nocturnal gecko species from the sandy sheets will segregate and dietary niche to allow for co-existence (avoidance of competitive exclusion); in case of the two species inhabiting gravel plains, we expect that they will have highly similar diets (competitive exclusion).
- We expect that overgrazing will have strong negative effects on gecko species composition, abundance, as well as effects (positive, neutral or negative) on dietary, microbial and parasite composition.
Identifying the colony role of late instar larvae in Psammotermes hybostoma: A potentially new sterile worker caste?
Identifying the colony role of late instar larvae in Psammotermes hybostoma: A potentially new sterile worker caste?
Project Summary

Termites, particularly the lower termites, have extraordinary developmental plasticity which has allowed them to adapt to varied and even harsh environments. Psammotermes hybostoma exemplifies this due to their capacity to survive in ecosystems with net zero productivity, such as arid deserts within the UAE.

This species also has a unique developmental pathway leading to older larvae which have an unknown function but are seemingly five times the size of the other individuals within the colony. It has been hypothesized that these individuals may produce metabolic water to help hydrate the colony, therefore could potentially be a new sterile worker caste. By collecting this species from the wild and undergoing behavioral experiments in the lab, the true nature of these individuals can be discovered.

Research Objectives
- Locate and collect colonies of the termite species.
Methodology

I will find the location of workers outside the nest which will help identify the location of the nest. Once the nest is found, using a shovel, I will collect the whole colony into a safe container. If external workers are not easily identifiable, bait traps can be used to lure out the workers, helping with locating the nest sites.

Expected Outcomes
- I will need to locate the nest of this termite species and then collect the whole colony for transportation back to the UK.
Arthropod Diversity in Ghaf Groves at the DDCR
Arthropod Diversity in Ghaf Groves at the DDCR
Project Summary

Arthropods are frequently sampled using a passive collection method called malaise trap. Such a trap can collect both flying, crawling, walking and jumping arthropods. A similar trap is used for sampling arthropods aerially (e.g. in trees) called a SLAM (Sea Land Air Malaise). A preliminary study during October 30th 2017 till April 3rd 2018 at the DDCR in a Ghaf Grove showed clear differences in arthropod diversity and abundance between samples collected using a malaise trap and SLAM trap. The proposed study will ascertain which arthropods are found at elevation within a tree in comparison to those that are found from the ground upwards.

Research Objectives
- Seasonal arthropod diversity within DDCR ghaf groves.
- Arthropod faunal difference collected by the two different methods (malaise and SLAM trap).
- Arthropod micro-habitat utilization within a ghaf tree.
- Suitability and refinement of the use of malaise and SLAM traps in a desert ecosystem.
Methodology

For direct comparison of data, two malaise traps and SLAM traps will be set up, one set in the ghaf grove situated in the Eastern region of the DDCR and the other in the Western region ghaf grove. The traps will be set up on October 2nd 2018 and will remain in situ for one year, with a possibility of extending this time period, depending on sampling success. Once operational, the sample bottles are exchanged every two weeks. Samples are passively collected into 70% ethanol and subsequently stored in containers with the same concentration. Samples are sorted in the lab into order, family, genus and species if possible, and where necessary, samples will be sent to a network of experts to help with identification. For population studies, individual species (whether identified or not) will be counted. Comparison will be drawn between taxa found in each trap and further literature review will take place to ascertain habitat requirements/feeding habits. For species where this information is not known, further research plans will be made.

Expected Outcomes
- Species lists.
- Data reflecting differences in arthropod populations from malaise and SLAM traps in a ghaf grove.
- Data collection on the efficacy of these types of traps in a desert ecosystem.
- Peer-reviewed journal publication; conference poster/presentation
Archaeological – Geo-archaeological Research in the DDCR
Archaeological – Geo-archaeological Research in the DDCR
Project Summary

Sorbonne University Abu Dhabi have the interest to develop an archaeological and geoarchaeological research at the DDCR broader area. The department of Archaeology and History of Art has a great experience in archaeological prospection in archaeological sites in UAE (Masafi, Sarouk Al Hadid, etc.) as also abroad - in Lebanon (Baalbek, Bisri, Chhim, etc.), Syria (Palmyra, Apamea), Greece (Delos) as well as in methods and technics related to photogrammetry and 3D scanning of artefacts and buildings. The Department of Geography and Planning has a great experience in geoarchaeological research till 2003 in Southern Europe (Balkan peninsula, Greece, Cyprus, France etc.), in Middle east (Egypt, Lebanon, Kingdom of Saudi Arabia, Kuwait, Oman, Bahrein) as in the UAE (Sarouk al Hadid, Masafi, Marawah, Ras Ghumais etc.) as well as in methods and technics related to the GIS, remote sensing and multi proxy analysis of the sediments and landforms.

The aim of this research proposal is the systematic archaeological field investigation in selected areas of DDCR, collection archaeological objects, classification of archaeological findings, and age determination, their localization by GPS, and their insertion in GIS database for evaluation and storage. Geomorphological mapping using remote sensing technics and fieldwork including sediments sampling for multiproxy analysis and dating.

The results of archaeological analysis and from the geomorphological and remote sensing analysis with the sediments multi proxy investigation can provide to us information’s about the potential existence of an archaeological site in the DDCR protected area.

Research Objectives
- Systematic surface Archaeological investigations using GPS and GIS.
- Classification and age determination of the archaeological findings. Possibility to conduct virtual analyze of 3D objects (non-invasive approach).
- Geomorphological mapping and spatial distribution of the archaeological findings.
Methodology

Systematic surface survey of the proposed area at the map. Four to five bachelor students of the department of Archaeology and History of Art of SUAD under the supervision of the Professor Dr. Ingrid Perisse Valero and another professor or archaeologist from Sorbonne University (name TBC) will do field investigations for three weeks in total on the area of interest.

Archaeological findings and artifacts field collection, localization by portal GPS and classification, digitalization of chosen artefact for 3D models. All the field data will be stored and processed in GIS database.

A report preparation of the findings and the localization of the artifacts and they propose the next steps of the archaeological investigations.

At the same time, one student, under the supervision of Prof. Dr. Kosmas Pavlopoulos and Dr. Haifa Ben Romdhane, will map the area of interest, the landforms, the sediments and they describe the geomorphological processes of the area. This research survey needs 2 weeks in total.

During the geomorphological/geological research a radar satellite data will be analyzed and process (INSAR, SAR, Sentinel, ALOS-pulsar2) to detect the substratum rocky paleo relief below the dune deposition. Dr. Haifa ben Romdhane and Prof. Kosmas Pavlopoulos will be responsible for this stage and deliverables.

Expected Outcomes
- Maps with the details position of the archaeological findings, including the written report of the archaeological findings classification report.
- Geomorphological map of the broader area of research. Analysis of the satellite images and interpretation. Report including the geomorphological and remote sensing results.