The Atriplex genus has several plant species that grow in ecosystems with high salinity, scarce humidity and high temperature ( Ramos et al. , 2004 ). These shrubs have been used for erosion control and rangeland rehabilitation in salt-affected and degraded areas ( Rani et al. , 2013 ; de Souza et al. , 2014 ). These shrubs are also utilized by farmers in dry ecosystems as maintenance feed for livestock during the drought feed gap ( Norman et al. , 2004 ; Pearce et al. , 2010 ) or for landscaping purposes ( Panta et al. , 2014 ; Ventura et al. , 2015 ).

Competition among plants in arid environments favors species that are able to become established in harsh environments and are able to reproduce and expand in the ecosystem. Plant species with higher tolerance to salinity, drought, and high ambient temperature are more successful to reproduce in low-resource environments ( de Souza et al. , 2012 ). Atriplex shrubs have adaptations enabling them to tolerate the adverse effect of salts internally or excrete salt from cells and tissues ( Flowers & Colmer, 2008 ). Therefore, they have an advantage over forage species unable to deal with salt in the soil and are therefore good competitors in arid zones with saline soils and can provide a supplementary feed source for livestock and wildlife under arid and semi-arid conditions ( Aganga et al. , 2003 ; Norman et al. , 2010 ).

Many species of Atriplex are good forage resource for livestock because of their adequate crude protein content and high dry matter digestibility ( Pinos-Rodríguez et al. , 2007 ; Mellado et al. , 2012 ); however, animals must combine the ingestion of A. canescens with more nutritious understory species because total dietary intake is limited by the salt content of A. canescens as animals ingest large amounts of salty shrubs ( Masters et al. , 2005 ).

Atriplex canescens occurs in many different arid ecosystems of the Chihuahuan desert and represents a significant source of forage for small ruminants, cattle, and wildlife ( El Shaer, 2010 ; Mellado et al. , 2012 ). In these ecosystems, the salt content of soils can mitigate the negative effects of water stress. For instance, plants in drying soils tend to thrive better in saline than in nonsaline soils, because salt-stressed plants grow not as much as those growing in nonsaline soils, therefore, they drain soil moisture slower than nonsalt-stressed plants.

Currently, little information is available on the effect of salt concentration of soil on the biomass production and nutrient content of A. canescens under field conditions. This knowledge is important to understand its ecological adaptations and provide valuable guidance on the exploitation of rangelands with mild concentrations of subsoil salt. The present study addresses the following questions: (1) what effect does salinity have on biomass production of A. canescens , (2) does nutrients of this shrub change in response to salinity Our working hypotheses were: increasing salinity levels would have a positive effect on biomass production; the magnitude of the soil salinity affects the nutrient content of this woody shrub. Our objective was to quantify the canopy cover and biomass production of A. canescens on a moderately productive clay loam site. An additional objective was to assess the effect of degree of salinity of soil on the nutrient content of A. canescens .

Despite fairly similar number of plants per paddock ( p >0.05) for sites with low, moderate and moderately high soil EC readings, A. canescens canopy cover was about two times higher ( p <0.01) in the paddocks with the highest soil EC readings compared to sites with lower EC of soil (Table 1). No statistical difference was detected in canopy cover between paddocks with low and moderate EC of soil.

Data of the 20 paddocks showed an average of 1401 A. canescens plants greater than 1 m of height/ha, with no significant difference between sites with low, moderate and moderately high readings for soil EC. A. canescens in paddocks with soil EC greater than 1.6 mS/m displayed significant increases in biomass (DM), when compared to paddocks with soils with EC<1.4 or 1.4 to 1.6 mS/m (Fig. 1). Likewise, DM of leaves of A. canescens was significantly higher in paddocks with values of soil EC greater than 1.6 mS/m.

A. canescens biomass production showed a significant positive correlation with values of soil EC (Fig. 2). Ash content of foliage of A. canescens increased with increasing soil EC but these differences were not statistically significant. Crude protein of A. canescens foliage ranged from 12.4 to 15.7%, with no differences between values obtained from plants growing in soils with low, moderate or moderately high soil EC readings (Table 2). NDF and FDF levels in A. canescens foliage averaged 57.3 and 31.4%, respectively, with no differences between A. canescens growing in sites with soils varying in EC values. No significant effects were observed for soil EC readings regarding percentages of IVDMD of A. canescens foliage. Levels of soil salinity did not modify the macro and microelements content of A. canescens 's foliage (Table 2).

A notable feature of the present study was that the aerial cover of A. canescens was stimulated by increased values of EC of soil. Changes in aerial cover can be influenced by two processes: recruitment of new A. canescens plants or coalescing of expanding A. canescens patches. In the experimental area, the number of plants of A. canescens was very similar among paddocks of different EC of soil which suggests that a greater development and expansion and possibly an increased number of stems per plant of A. canescens occurred with the highest EC of soil, which lead to a considerable coalescing of this shrub. A greater A. canescens canopy cover affects both vertical and horizontal water fluxes, therefore, the greater expansion of A. canescens in slightly saline soils makes it a particularly prized fodder shrub to improve soil and water conservation.

As soil EC increased, the likelihood of greater biomass was enhanced. This is in line with results of Gomez-Silveira et al. (2009) and Pan et al. (2016) where the growth of Atriplex species seedlings was enhanced by moderate salinity and unaffected by high salinity. This positive association has also been reported for other Atriplex species ( Redondo-Gómez et al. , 2007 ; Bouchenak et al. , 2012 ; Tsutsumi et al. , 2015 ). The mechanisms promoting the growth improvement at moderate salinity levels are not fully understood. In halophyte plants, a moderate increase of soil salinity leads to a transient increase of stomata number, maximum opening area and size, which is an adjustment to a particular demand of CO ₂ ( Shabala et al. , 2012 ; Koyro et al. , 2013 ). Carboxylation capacity does not seem to be adversely affected by salinity in Atriplex portulacoides , but it is stimulated by moderate salt concentrations which contribute to increased growth rates.

Additionally, a moderate increased salt content of soil seems to enhance photosynthetic capacity and consequently the growth of A. canescens . Pan et al. (2016) suggest that Na ⁺ promote the C ₄ photosynthetic process of A. canescens plants and consequently improve the water use efficiency under high levels of soil salinity. Another mechanism for xero-halophytes to increase growth rate is the development of a greater photosynthetic area, allowing its fastest growth to occur at various soil salinity levels ( Redondo-Gómez et al. , 2006 ). Additionally, moderate soil salinity led to an increase in the water use efficiency in halophytes ( Koyro, 2006 ; Redondo-Gómez et al. , 2007 ; Hussin et al. , 2013 ).

Nutritional status of A. canescens relative to readings of soil EC did not show any significant change. Total soluble carbohydrates and total soluble protein in leaves of A. nummularia are increased as water salinity increased ( Hussin et al. , 2013 ). The divergence between the aforementioned study and the present trial derives from the fact the former experiment was carried out in hydroponic greenhouse system and the later was conducted in the field with much narrower salinity levels. Results of the present study suggest that photosynthesis, biochemical and photochemical capacities of the leaves of A. canescens in slightly saline soils were not impaired.

A. canescens met the protein and energy needs of livestock in different physiological stages and is considered adequate as forage at any stage of development ( Pinos-Rodríguez et al. , 2007 ; Mellado et al. , 2012 ). Likewise, the foliage of A. canescens presented adequate digestible energy (based on IVDMD), and concentrations of all mineral elements for grazing farm animals and wild ruminants ( Suttle, 2010 ), although adequate water at low salinity levels would also be necessary since A. canescens contains high quantities of sodium and potassium. Copper supplementation would be advisable for animals foraging this rangeland with predominance of A. canescens , as this element fails to meet the Cu needs of grazing ruminants and could cause copper deficiencies in livestock and wildlife that do not have access to copper supplements or other forages high in copper concentration.

The IVDMD range of A. canescens at all sites fell within the range (34.3 to 80%; different digestibility techniques) for DM digestibility noted for different Atriplex species ( van der Baan et al. , 2004 ; Pinos-Rodríguez et al. , 2007 ; Askar et al. , 2016 ). The NDF content of leaves of A. canescens was higher that values found by other researchers ( Ben Salem et al. , 2004 ; Otal et al. , 2010 ). As NDF is closely associated with feed intake, these data suggests that the moderate NDF values of the leaves of A. canescens in this trial would allow adequate intakes by small stock and cattle.

A. canescens would be a good choice to consider in revegetating rangelands because it offers substantial quantities of forage and nutritional qualities generally adequate to meet requirements of livestock and wildlife, although high ingestion of this plant material with elevated salt content requires increased water intake and could depress feed intake.

In the present study the ash content of leaves of A. canescens ranged from 14.5 to 16.4, values lower than those reported for other Atriplex species, where ashes exceed 20% of the DM ( Khan et al. , 2000 ; Abu-Zanat et al. , 2003 ; Meneses et al. , 2012 ). The mineral composition of A. canescens appeared adequate to meet the requirements of small ruminants, cattle and wild herbivores grazing such types of forage ( Suttle, 2010 ).

Leaf sodium content increased with increasing soil EC values, but this difference was not statistically significant. This is contrary to other studies where sodium concentration in the foliage of Atriplex increases significantly with increasing soil NaCl concentrations ( Benzarti et al. , 2014 ). However, most studies where the association between soil salinity and sodium content of Atriplex plants have been carried out in pots (inert material) with the use of complete nutrient solution added with different salinities. Likewise, Na ⁺ and Cl ^- in leaves did not decrease K ⁺ . These results indicate that A. canescens in the site of the present study was just slightly challenged with salinity which did not result in salt-induced nutritional imbalances.

Cations (Ca, K, and Mg) concentrations did not decrease with an increase of soil salinity and levels of these minerals were adequate and constant throughout the EC values in the soil, which does not agree with results found for other halophytes ( Ayala & O'Leary, 1995 ; Khan et al. , 2000 ; Belkheiri & Mulas, 2013 ). However, most studies where increased levels of NaCl have induced decreases in Ca, K and Mg in halophytes have been carried out in plant growth chambers and results under controlled conditions normally does not translate back to field situations ( Poorter et al. , 2016 ).

In summary, results of the present study indicate that, under field conditions, A. canescens is strongly stimulated by soil electric conductivity values >1.6 dS/m, which suggests that, in drying soils, the accumulation of moderate levels of salt would alleviate rather than exacerbate the stress load on this fodder shrub. Additionally, these results show that nutrient contents of foliage of Atriplex canescens are not adversely affected by slight salinity stress in an arid climate.