The main drivers of biodiversity loss relate to the rapid expansion of settlements combined with the interaction of economic, social and cultural factors, and the increasing occurrence of extremely high temperatures and uneven rainfall, with endemic vegetation at risk due to the frequent planting of mono-plant cultures. (Abuzinada et al., 2005) Changes in agricultural practices in the southwestern region of the country relate to the enlargement of farm areas, has also led to the loss of agricultural terraces. This in turn has resulted in the loss of microhabitats for plants and increases soil erosion and flash flooding. Subsequently as people relocate to urban environments, there is also a loss of traditional knowledge relating to traditional agricultural practices, and understandings of ecosystem management. The gathering of naturally occurring foods and medicines is still important to the people and economy of the country.

Additionally, recognizing the importance of plant habitat in cities is an important criteria for selecting sites for conservation action. To date the majority of research on mosaic landscapes has focused on the loss of natural habitats and/or their fragmentation rather than on the construction of new habitats in extremely damaged or at risk ecosystem ecologies. Investigating the spatial range of simple to complex constructed habitats that may be sustained in extreme ecology contexts such as those found in the Arabian Peninsula, constructing urban garden habitats will generate a natural diversity and through scaling of these initiatives over time, will enable larger habitat formation for at risk species. (Goddard et al., 2010) Constructing habitats in cities also will contribute significantly to the development of ecosystem services. At the present time, exploring the construction of urban garden habitats for biodiversity has not been incorporated into the Saudi Arabia's biodiversity conservation action plans.

The city of Arriyadh has actively engaged in establishing new habitat, but a number of key challenges face regional and local stakeholders, the primary being the absence of design research and guidelines on how to establish and sustain biodiverse ecological patches in extremely arid ecosystems. There is a need to define the specifications of habitat types to be constructed, in addition to establishing guidelines for the design and construction of ecological corridors and their connectivity range. Policies are also needed at the local governmental level administration and maintenance of neighborhood community garden initiatives. There is also an absence of assessment metrics for arid constructed habitats specific to urban areas in the Arabian Peninsula.

Urban gardens enable people to interact with nature and establish ecosystem services that impact the quality of urban life. The management of urban gardens alters land cover and species diversity, with groups of gardens forming green patches that become important refuges for natural species. Information on wildlife gardening is limited in ecological research, with the majority of research undertaken in developed countries. (Goddard et al., 2010) Urban habitats are constructed with varying types of vegetation or landcover to form a distinctive environmental condition, providing natural resources required by different species or species group. (Wang, 2008) The design of biodiversity enabling habitats is related to the need to increase the abundance of birds, butterflies, and other animals and plants valued for their aesthetic and functional attributes. (Hough, 2004) Numerous challenges face municipalities seeking to balancing recreational and ornamental need with the construction of greenspace that will sustain and increase the density of native biodiversity. Different spatial arrangements at macro and micro scales enables the formation of heterogeneity, an important structuring consideration for sustainable biodiverse habitat formation. (Goddard et al., 2010) Understanding the relationships between ecological function and spatial structure and scale will also enable the ecological consequences of proposed spatial solutions or risks to be anticipated and modeled. [(Leitao & Ahern, 2002); (Pace et al., 2015)] To date there is a limited understanding of how constructed spatial habitats relate to ecological processes, species establishment and dispersal, and species responses in extreme arid climates of the Arabian Peninsula. Arid urban environments pose challenges to constructing habitats for biodiversity due to the absence of water, good soil types, present vegetation range, and a symbiotic background mosaic. In extreme arid urban contexts, spatial patterning of habitats and their patches as discrete or integrated assemblages may expose constructed habitats to higher risk and need for ecoservice management due to the type and patterning of land cover. [(Landis, 2017); (Pace et al., 2015)] The identification of ideal urban sites and neighborhoods to establish urban garden initiatives requires study of the site solar exposure and above variables, but also depends upon the symbiotic commitment of local community caretakers to sustain the urban gardens over time, and the municipalities design expectations of constructed biodiverse habitats as compared to ornamental or recreational urban landscapes.

One of the central concerns of landscape ecology is how spatial heterogeneity shapes the dynamics of ecological systems and temporal processes. (Pickett & Cadenasso, 1995) Increasing the heterogeneity of landscape ecologies depends more on the spatial structuring of patches and habitats rather than the spatial range of scales. (Pickett & Cadenasso, 1995) Different approaches to the spatial structuring of constructed habitats have been found to limit the ability of the habitat to expand over time, resulting in a stigmed resilience to rapidly changing ecosystem characteristics. (Pace et al., 2015) Research suggests that urban habitats create a consistent set of habitat templates (hard-surfaced environments with shallow compacted soil and extreme hydrological conditions and temperatures) that tend to attract species drawn disproportionately from rocky habitats [(Larson et al., 2000); (Larson et al., 2004)] Urban landscaping practices in Saudi Arabia focus on imported monoculture vegetation acclimatized to the extreme arid and solar environment. (Abuzinada et al., 2005) Different spatial habitat formations represent different conditions that attract different flora original or imported, and distinctive vegetation composition. (Lundholm & Marlin, 2006) Different taxa will perceive and respond to landscape structure at many spatial scales depending on a range of parameters. A key challenge is to maximize vegetation complexity throughout the ecosystem. (Goddard et al., 2010)

Constructed habitats display variations in how they assemble patches for different species of flora and fauna. Studies of spatial heterogeneity of a habitat frequently link to the functional ability of the habitat to support and generate variety, richness and abundance of living organisms. (Landis, 2017) Describing the morphology of spatial patterns in constructed habitats (urban gardens) requires a fine grain analysis of patch and sub-patch, in order to understand how constructed habitats form resiliency and change over time. Studying spatial patterns reveals regularities in underlying ecological processes, and function of ecological parameters within them. The fine grain patterns present in urban gardens may enable them to enhance biodiversity due to more frequent and diverse disturbances in the patch system. (Pickett et.al., 1995)

Essential characteristics of natural and constructed habitats are their assemblage of spaces (patches and sub- patches), and shape (geometry, size, vertical height). Assemblage complexity enables the formation of a larger more intact habitat area. The integration of vegetation patterns of different taxa, different specie, maturity range, with a diversity of spatial assemblage approaches will provide richer more dynamic habitat formation that is more resilient to ecosystem change. The size and shape of patches has been shown to determine to a large degree their ecological and functional characteristics with patch spatial structure, described as homogeneous or heterogeneous. [(Pickett & Cadenasso, 1995); (Landis, 2017); (Pace et al., 2015)] The connectivity of patches is an additional essential structure that enables the movement of flora and fauna across patches and meta-zones. (Leitao & Ahern, 2002) Corridors that form connecting linkages allow both the dispersion of flora and fauna throughout the urban system across spatial scales. In contrast discrete design approaches exhibit strong edges, boundaries or limits with no apparent connectivity, that inhibit the formation and expansion of bio-diverse ecosystems. Discrete patches may be differentiated by biotic and abiotic structure or composition. (Pickett & Cadenasso, 1995) Edges or overlapping formations are characteristics of shape that have been associated with the spatial dimension of habitats and ecological processes. (Landis, 2017) Constructed habitats have been found to reshape the existing microclimate profile of a site by the organization of vegetation patterns to reduce the lands exposure to solar radiation and high winds.(Lin et al., 2018) Three types of landscape function are associated with the shape and complexity of landscape habitat patches; Production such as food, wood, recreation and community services; Protection natural functions maintenance and enhancement of biological species; Regulation – microclimate formation, temperature, wind and air quality. (Leitão et al., 2012) The minimum habitat requirement for an indicator species has generally been identified to be in the area of at least 0.3-0.5 hectares. [(Redpath, 1995); (van Apeldoorn et al., 1992); (Harris, 1984); (Robbins et al., 1989); (Rudd et al., 2002)]

In arid environments the territorial range may be substantially lower as a single tree may form an important mediating link in a stressed ecosystem. Microclimates result from the interaction between prevailing climate and objects in the landscape. (Brown & Gillespie, 1995) By focusing on the solar path orientation around the site vegetation placement in gardens may construct shading screens and canopies to decrease solar exposures of the garden sites. Tree shading is an important mediator for bioclimatic formation. A study of olive tree shadow on soil temperature during summer period can reduce the temperature by 11° C. (Panagopoulos 2007). Urban green structures cool hot air by evapotranspiration, provide shading to the ground and adjacent building walls, and reduce the velocity of wind, regenerate air quality, absorb air borne dust and filter noise.

The aim of this study is to describe the spatial heterogeneity composition of two recently established urban garden habitats, and describe their spatial formation.

Saudi Arabia contains three distinct climate area’s, coastal area’s, internal dry area’s, and highlands. (Alrashed & Asif, 2015) In Jeddah, a city on the Red Sea coast, residents experience an average temperatures of 32° C to 49° C, with 45° C in the shade during summer months. In winter, average temperatures range from 24° C – 30° C. Relative humidity varies seasonally from 55% - 65% with significant levels of air pollution and airborne dust. Vegetation in the Makkah region and city of Jeddah is generally sparse with 60% in low-lying areas. Cool high mountains, arid deserts and steeps, and hot semi-arid coastal plains characterize Saudi Arabia’s natural environment. Dry hot winds are especially threatening to bird and plant resilience during summer months. Low and unpredictable rainfall, and severe droughts pose challenges to conserving the kingdoms biological diversity. Infrequent rain primarily falls in the southwestern escarpment on the Red Sea coast.

This visual survey of Jeddah urban gardens, establish by private individuals on vacant land in a new urban district, examine how the gardens are spatially formed to construct patch habitats for a range of vegetation species. Two constructed habitat (CH) nodes are distinguished as Site 1 (S1) and Site 2 (S2). Patch formations within the node are identified as Patch 1,2,3 (P1, P2). Assemblage heterogeneity of the spatial morphology of the garden is coded to

(A) describe the dimension of sub-patch, (B) describes patch core dimension, (C) describes the edge position. Site position in the connectivity matrix is shown in dark grey. A limited description of the taxa classification is made.