Nurse‐Assisted Ecological Restoration Methods In Disturbed Environments

As nurse plants are shown to enhance soil fertility and decrease water or temperature stress, they could prove to be an effective tool in increasing revegetation success in areas where seedling emergence and survival is constrained by inhospitable abiotic environments, severely degraded habitat, or by herbivory. Traditionally, reforestation techniques involved removing pre-existing woody vegetation growing in close proximity to newly planted trees since they were considered potential competitors (Rey, Siles, & Alcántara 2009). For instance, studies have shown that competition may be severe in drought prone ecosystems (Davis, Wrage, & Reich 1998; Saccone, Delzon, & Pagès 2009). It has since been argued that such techniques may be suboptimal if the objective is to restore natural plant communities (Gómez‐Aparicio, 2009; de Toledo Castanho & Prado, 2014).

There is an ever growing body of evidence which illustrates the importance of spatial proximity among plants. The application of nurse plants in forest restoration is well studied in the Mediterranean Basin (Castro, Zamora, & Hódar 2002, 2004; Gómez‐Aparicio et al. 2004). Most forests in the Mediterranean area have been heavily modified by millennia of human intervention and has environmental conditions, characterised by being highly variable in both time and space (Gomez-Aparicio et al. , 2004). Furthermore, previous reforestation attempts have been thwarted with extremely high rates of early plant mortality (Rey, Siles, & Alcántara 2009). This makes the Mediterranean area an ideal environment to test the feasibility of nurse-assisted restoration methods. Experimental studies by Castro, Zamora, & Hódar (2002) found that the survival of two native conifers was substantially higher when planted under individuals of the shrub Salvia lavandulifolia.

This study area was revisited in 2004 where it was determined that the reduction in solar radiation underneath the nurse canopy is likely to be the primary factor facilitating the survival and growth of the native conifers (Castro, Zamora, & Hódar 2004). Other studies have found the strength of nurse effects varies in different ecological settings in the Mediterranean mountains (Gomez-Aparicio et al. , 2004). By contrast other studies in the Mediterranean Basin found that the strength of the positive interaction varied on a spatial and stress gradient (Gomez-Aparicio et al. , 2004). For instance, Gomez-Aparicio et al. , (2004) found that nurse shrubs have a stronger facilitative effect on the growth and survival of seedlings grown at low altitudes and on drier slopes compared to those grown at high altitudes or on shady, wetter slops in Mediterranean areas (Gomez-Aparicio et al. , 2004). Thus, nurse assisted restoration is more relevant in open areas, where there is higher temperature and irradiance (Gomez-Aparicio et al. , 2004).

This highlights the importance and management implications of environmental variability in plant-plant interactions. The majority of research focused on nurse-assisted restoration is conducted under moderate drought stress conditions in semi-arid ecosystems (Gómez‐Aparicio et al. 2004; Rousset & Lepart, 2000). There is an apparent lack of information regarding the use of facilitation as a tool for restoration in arid and desert ecosystems. Recent studies testing the use of nurse plants in arid ecosystems found contrasting results from restoration studies conducted in semi-arid ecosystems (Jankju, 2013; Noumi, Chaieb, & Michalet 2015). For instance, leguminous shrub Astragalus gossypinus has a facilitative effect on target species during moderate water stress in spring, but shifted to a negative effect during severe summer drought conditions (Jankju, 2013). It is thought that competition for water and nutrients is more important under severe drought conditions (Jankju, 2013). Similarly, a recent study found that the use of chenopod shrub and a perennial grass species as nurse plants in arid grazed ecosystems is not a suitable restoration tool in this systems (Noumi, Chaieb, & Michalet 2015). It is believed that only available target species in this system (Acacia tortilis) cannot benefit from a nurse plant association since it is too shade-intolerant and highly drought tolerant (Noumi, Chaieb, & Michalet 2015).

Together, these findings highlight that conclusions drawn from experiments conducted in semi-arid areas cannot be applied to all water-stressed ecosystems. Future studies should investigate the use of other nurse plant species in these systems. In particular, leguminous shrubs should be used to assess if facilitation can be observed in these systems, since they are well known as effective nurse plants relative to grasses or chenopods (Cuesta, Villar‐Salvador, & Puértolas 2010). It is now apparent that the direction and importance of plant interactions during ecological restoration depends on a suite of factors (Noumi, Chaieb, & Michalet 2015). This can also explain inconsistencies within the literature. There is experimental evidence that nurse-assisted restoration can provide an effective tool in certain ecological settings (Castro Castro, Zamora, & Hódar 2004), whereas other ecosystems prove to be unsuitable (Jankju, 2013; Noumi, Chaieb, & Michalet 2015; Gomez-Aparicio et al. , 2004).

Nurse plant selection needs to be based on the species traits and strategies of both the nurse and the target species in order to determine the success of a restoration project. The development of a robust reforestation project requires case-by-case application. Conclusions(1) Current knowledge is heavily biased towards testing the direct nurse-plant interaction. Research on the indirect interactions, such as protection against herbivory, seed trapping, facilitation of pollinator visits, and the role of below ground soil communities, is less frequently performed. It is critical that future studies aim to decouple the mechanistic pathways of the nurse-plant syndrome in order to better understand the ecology of communities defined by nurse-plant interactions. (2) There is a general focus on the role of nurse plants under mildly stressful conditions, namely semi-arid environments. Little attention is paid to extremely harsh environments, such as arid regions, or to seemingly favourable ecosystems, such as temperate or tropical plant communities. In light of the rapidity of climate change, research should be interested in identifying species that may act as nurses. (3) The importance of nurse-plant interactions across stress gradients remains heavily disputed among scholars with numerous studies leading to contrasting conclusions. Recent studies highlight the fact that the strength of a positive nurse-plant interaction is determined by a wide range of different factors and explain contrasting findings. Caution should be taken when studying nurse-plant interactions with all factors taken into consideration before making conclusions. (4) There is experimental evidence that nurse-assisted restoration can provide an effective tool in many ecological settings.

However, the success of a nurse-assisted restoration project depends on a suite of factors, namely species traits and strategies of both the nurse and the target species, the annual rainfall, resources verses non-resources stress factors and biomass of potential nurse plant species. Recent studies have highlighted the importance of evaluating these factors when assessing the suitability of facilitation as a relevant tool for ecological restoration at any given site. As nurse-assisted reforestation approaches are shown to promote rapid revegetation, minimise cost and effort, with low impact on plant communities, the use of nurse plants should become a major environmental goal worldwide. There is an urgency to gain a better knowledge on the application of this method as climate change will challenge the survival and growth of seedlings in many ecosystems around the world.