The potential for desert plant species to capture carbon (C) has not been well investigated, especially with the potential impact of enhanced greenhouse gas emissions. This was supported by leading investigators such as Lal and colleagues ¹ , who stated that the contributions of soil organic C dynamics on CO ₂ as a greenhouse gas, and its strategic importance in mitigating enhanced atmospheric CO ₂ are not well accepted. Moreover, estimates of net ecosystem production in deserts are incompatible with recent net primary production and C pools in deserts ² . The potential for desertified ecosystems to sequester C are still considerable ³ . But a significant part of C can be sequestered within a reasonable time frame ³ . Especially that estimates of soil organic carbon (SOC) pool in desert and semi-desert ecosystems, for instance, are in the order of 159-191 billion tons C. But the role of SOC on historic increase in atmospheric CO ₂ are not widely recognized ³ .

Li et al. ⁴ studied soil carbon sequestration potential in semi-arid grasslands with significant offsets to increasing atmospheric CO ₂ levels may be achieved in addition to erosion control and improved wildlife habitat. Atmospheric CO ₂ concentrations continue to rise due to human activities. CO ₂ is a leading greenhouse gas, the levels of which are directly affected by anthropogenic activities. In forested ecosystems there is a ubiquitous initial increase in net primary productivity ⁵ , however, as these systems shunt more C belowground heterotrophic communities may increase activity, resulting in net soil C losses. Similarly, in grassland ecosystems C uptake may increase, which in turns leads to increased rates of C cycling belowground and speeds up the loss of C from the slowest cycling C pool ⁶ . There is a need for C storage in different pools, with long residence times or which transfer C from organic to inorganic forms. Arid and semi-arid regions, for instance, account for about 30% of the earth terrestrial surface ⁷ and these important areas are expanding due to anthropogenic impact. However, these arid environments have been understudied, despite recent studies that have shown these environments may be sinks for atmospheric CO ₂ ⁸ . The stability of soil C is also affected by desertification processes ⁹ .

There is a strong need to study semi-arid shrublands and deserts, since studies have shown that they remove similar amounts of CO ₂ from the air as forests and grasslands ^{10– 13} . These semi-arid lands have high CO ₂ uptake due to three main factors: 1) the expansion of shrubland vegetation, 2) the cryptobiotic crust communities and 3) the accumulation of C in soil carbonates. These mechanisms may increase C in both above and belowground C pools which may sequester C and retain it in pools with various mean residence times. It has been estimated that arid and semi-arid soils may remove as much as 5.1 g C/m ²/year from the atmosphere ¹⁴ . Stanbery et al. ¹⁵ examined variability in soil carbonate storage at the pedon-scale, quantified analytical and measurement error isn soil carbonate measurements and defined the amount of carbonate stored in gravelly vs. non-gravelly soils.

Other studies have shown that this accumulation of C in soils as carbonates can occur over a very short period of time ¹⁶ and that the conditions that lead to increased soil carbonates may be controlled by vegetation and the availability of calcium and bicarbonate ¹⁷ . The use of native plant species has also been suggested to improve soil C sequestration ¹⁸ . Land management may affect future amounts of SOC in semi-arid areas thereby turning them from sources into sinks of C ¹⁹ . Encouraging sustainable land management practices can contribute to restoring soil organic C stocks and to creating new C sinks, leading to improved C sequestration potentials ²⁰ . Additionally, land management may turn sources of SOC into sinks of C in semi-arid areas ¹⁹ .

The present work was intended to consequently contribute to assessing C pools in the hyper-arid ecosystems of the UAE, using biomass that affects C allocation in the belowground compartment. Of great interest was to determine which species present the greatest increases in belowground organic and inorganic C pools. Globally, soils store more than twice the amount of C than is present in the atmosphere and terrestrial biomass C pools, combined ²¹ . Soil organic C may also have a longer mean residence time than atmospheric CO ₂ (5 years) or terrestrial biomass C pools (9 years, globally averaged) ^{22, 23} . Even more exciting is the ability of some arid ecosystems to take atmospheric C, fix it via photosynthesis, move that organic C belowground and eventually transform that organic C into inorganic C under relatively short time scales ⁸ . These arid ecosystems have not been well investigated and it is unclear which plant species possess the greatest ability to sequester C both below as well as above-ground. In the present study we measured C sequestration potentials of different soil/vegetation associations as affected by the dominant tree and shrub species. It is therefore aimed to accurately quantify the amount of C removed from the atmosphere and stored in above and belowground C pools of plants that are native to this hyperarid environment.

Here we measured the potentials of C pools in different land forms (soil/vegetation associations) as affected by the dominant tree and shrub species. Plant characteristics regulate soil carbon storage through the outcomes of C assimilation and its storage belowground ³¹ . Changes in ecosystem soil C may be in response to a variety of management as well as environmental factors ³² . It therefore aimed to accurately quantify the amount of C removed from the atmosphere and stored in above and belowground C pools of plants that are native to the UAE arid environment. The C stored in aboveground components constitutes one-third of that in soils ³³ . However, this differs from ecosystem to ecosystem as a relationship between soil C and climate have been reported, especially the positive impact of precipitation ³⁴ . Mountain valleys sequestered the highest C in the phytomass compartment, with an average of 3.6 tons per hectare. It may be because of the limited anthropogenic impact on such ecosystems in the UAE. Human intrusions, such as that in relation to degradation, added to the further depletion of soil C ³ .

Desertification, degradation of soil and vegetation because of anthropogenic factors, among others, affect close to 4 billion ha globally ³⁵ . The rate of desertification is estimated at 5.8 million ha per year ³⁵ . Here it was the case for gravel plains, which was the most affected soil/vegetation association in the UAE. Unwanted plant species, which were indicators of overgrazing, were most abundant in such plains ³⁶ . The proportions of carbon sequestration was highest in the soil compartment within sand plains (i.e. 49.1%). Soil carbon sequestration was lowest for gravel plains, with an average of 28.3%. At the species level, the highest contribution was from Z. spina christi, with an average of 1.73 tons per hectare of sequestered C. A different Ziziphus was reported to sequester 24 Mg/Ha ³⁷ . For the introduced P. juliflora, we reported by far the highest average contribution of tons of C per hectare, most of which is in the aboveground part. It is important to note that such introduced species are widespread and still occupying more areas in the UAE. In semi-arid areas to which Prosopis and Acacia are adapted, 6.210 ⁹ Mg of carbon would be sequestered. Litter is an important aspect in the dynamics of soil C pools. Biological and physical processes play critical functions in litter decomposition ³⁸ . In this study, we report that the average C content of litter from A. tortilis was the highest at about 49%. A large accumulation of C in the litter layer was reported in a forest ecosystem ³⁹ . Land management also plays important roles in litter accumulation. Higher litter accumulation was reported when the land management was appropriate ⁴⁰ .

In short, some land forms, such as mountain valleys, were shown to sequester more C than others, which constitutes a good reason to improve their conditions and minimise anthropogenic impacts. Focusing future identification of protected areas can be a practical option to improve these types of soil associations. Few shrub/tree species were also reported to have high potentials for C sequestration in the hyper-arid environment of the UAE. Species such P. juliflora are an example of floral species that accumulate large amounts of soil C. Such species, despite its negative invasive features, may be a good candidate to re-vegetate many of the marginal lands in the UAE.