The Plants That Were Here Before the Plan
During the first heavy rain after a long summer, many of us stand at the window and watch the street. The water running past is not clear but dark brown, loaded with soil that minutes earlier belonged to a hillside, a field, or a riverbank. We treat the scene as passing, yet it is a silent loss that never returns. In the Mediterranean Basin, where about 67% of the land is classified as drylands according to the study by Nunes and colleagues, this scene is no exception. The paradox is that what stops this bleeding rarely comes from concrete or retaining walls. It comes from plants that lived here before the first plan was ever drawn. These plants bind the soil with their roots, slow the water, and trap silt, and some of them summon the rest of the forest to their side. But not all native plants are equal. Every context has a plant that suits it, and every slope has an orientation that governs its fate. So which species have proven successful, where, and why?
When Choosing a Plant Becomes an Engineering Decision
Designers tend to treat the planting list in a landscape project as a final layer of decoration, arriving after the serious matters of grading, drainage, and walls have been settled. The evidence gathered by researchers across the Mediterranean Basin reverses that order. In a survey of ecological restoration practitioners across the basin, Nunes and colleagues documented that European Union countries rely mostly on native species, which account for about 48% of seedlings and 47% of seeds. The study noted that using locally sourced seeds and seedlings is increasingly recognized as a condition for restoration success. The reason is that species adapted to a site’s soil and climate enter the network of local biotic relationships that an ecosystem needs to sustain itself and remain resilient.
This means that choosing a plant species is not a matter of aesthetic taste but, quite literally, a structural specification. An unsuitable plant does not die alone. It leaves behind an erosion channel, a collapsing slope, or a bank that wears away. Hence the question every landscape architect should ask: which plant for which function? A seasonal channel that is dry most of the year needs something different from a riverbank bordering irrigated fields, and a scorching south-facing slope differs from a moister north-facing one. Researchers have tried to answer this question with numbers rather than intuition, assigning plants scores that measure what their roots and stems do when water flows. By that same logic, we move to the place where everything begins: the seasonal channels we see only on the day they fill with water.
Roots Measured in Kilopascals
Seasonal channels and gully bottoms are among the most deceptive elements of the Mediterranean landscape. They look dry and still for long months, then a single rainy season is enough to turn them into a watercourse that digs and sweeps away. De Baets and colleagues developed a methodological framework for selecting species that restrain gully and rill erosion, applied it to a Mediterranean ecosystem, and scored plants according to their mechanical and hydrological traits.
Spiny rush (Juncus acutus) topped the ranking for preventing erosion caused by concentrated flow and for trapping sediment. It prefers moist environments and fine substrates, which makes it a first candidate for rehabilitating channel beds. Canary Island tamarisk (Tamarix canariensis) is a small tree whose root cohesion reaches 36.74 kilopascals, the measure of how much the roots add to the soil’s resistance to sliding and collapse. To this are added its strong resistance to bending under the pressure of water and its tolerance of seasonal floods. White broom (Retama sphaerocarpa) stands out with a higher root cohesion of 38.56 kilopascals and a deep reinforcement of the soil suited to stabilizing banks. Lygeum spartum is a salt-tolerant grass, and when its clumps thicken at gully outlets they trap the fine sediment arriving from upslope and keep it from entering the channels. Finally comes oleander (Nerium oleander), a shrub that earned the highest score on the stiffness index and withstands the floods of seasonal watercourses.
The design meaning is clear: species do not work alone but in sequence. The grass at the outlet traps silt, the shrub in the bed fixes the channel, and the deep-rooted one reinforces the bank. But what happens when the land beside the watercourse is a field irrigated and fertilized?
A Green Strip That Saved the River from the Fields
Riparian vegetation, meaning the plants that grow along the edges of watercourses, often does work that treatment plants cannot. Aparício and colleagues studied the Almonda Valley in Portugal and modeled the role of existing riparian cover in regulating the flow of nutrients and sediment from agricultural land. They found that the riparian strips already in place, many of them wider than ten meters, reduced nitrate and phosphorus exports from irrigated fields by about 90%. More important than the figure itself, the existing cover kept summer nitrate concentrations in the watershed from exceeding the thresholds set by the European Union Water Framework Directive.
This result reveals that the riparian strip is infrastructure working in silence, and that its width and density are design variables as important as pipe diameter in a drainage network. When the strip is cut away to widen a field or level a road, the price does not appear on the drawing. It appears in the quality of the water months later.
Among the plants that the De Baets framework identified for these zones is common reed (Phragmites australis), which combines moderate resistance to bending with a high capacity to obstruct sediment, and a notable root cohesion at depth of 35.18 kilopascals. It reinforces the lower layers and traps silt at the same time. Rosemary (Rosmarinus officinalis) stood out on abandoned fields, recording a high capacity to obstruct sediment (0.118 meters per meter) and good resistance to bending. It often settles in the concavities where gullies begin to form, cutting off their development. But erosion does not wait only at rivers, and its harshest forms appear on slopes.
Unforgiving Slopes: The Hillside’s Orientation Writes the Plant List
On Mediterranean slopes it is not enough to ask which species. One must also ask which direction. South-facing slopes are hotter and drier, while north-facing ones are moister and allow a wider range of choices. The De Baets framework described this contrast precisely. Esparto grass (Stipa tenacissima) is a highly drought-tolerant grass suited to steep south-facing slopes, and it protects against erosion from concentrated flow and against shallow slides. Its dense network of fine roots in the upper soil layer (zero to ten centimeters) forms a barrier that reduces erosion, earning three points out of four on that index. It is recommended in combination with Salsola genistoides, which received the highest score for slope stabilization and resistance to bending, and which regrows after fire or grazing. North-facing slopes suit species such as Brachypodium retusum, a rhizomatous grass that regenerates after fire and improves the cohesion of soil aggregates, and Anthyllis cytisoides, which earned the highest score for root cohesion. For saline environments, Mediterranean saltbush (Atriplex halimus) is suitable, and for severe slopes, Ononis tridentata, which earned the highest score for resistance to bending.
The picture becomes clearer still on roadsides, which are engineered ground par excellence. Bochet and colleagues conducted extensive experiments on semi-arid roadslopes in eastern Spain. They found that hoary plantain (Plantago albicans) germinates within only six days at a water stress of negative 350 kilopascals, making it one of the few species able to settle on the harshest south-facing cuts carved into the road. Drilling proved effective for establishing this species in compacted cuts. Cotton lavender (Santolina chamaecyparissus) also succeeded on south-facing cuts, though with lower survival, and orchard grass (Dactylis glomerata) was used in selected seed mixtures for hydroseeding. These experiments carry a message for contractors: compacted soil after excavation is not the end of the conversation with plants. It calls for a suitable establishment technique and a species that knows how to start from nothing. Yet not everything concerns what lies above the soil, so what about soil that has already been poisoned?
Soil That Carries the Memory of Metals
Near Marseille, centuries of industrial activity left soil burdened with arsenic, copper, lead, antimony, and zinc. After about a century of spontaneous recovery, Heckenroth and colleagues found that certain native plants had managed to settle there. They developed a non-destructive methodology to screen these plants and determine their potential for phytoremediation, the use of plants to treat contaminated soil. The most prominent is Coronilla juncea, a leguminous shrub whose presence correlated significantly and positively with mixed metal contamination, and which has been used successfully in restoration projects on both contaminated and uncontaminated soil. Globularia alypum, a pioneer shrub in plant succession, recorded bioaccumulation and translocation factors below one for arsenic, copper, lead, and zinc, meaning the metals remain locked in the roots and do not rise to the leaves. This makes it a candidate for phytostabilization of metals. Other tolerant species were also listed, such as Piptatherum caerulescens, Biscutella laevigata, and Silene vulgaris.
On mined land, Gairola and colleagues reviewed experiments in which native species were used, among them Acacia nilotica, Bermuda grass (Cynodon dactylon), common reed, and cattail (Typha latifolia), to restore fly ash dumps and mining sites.
The greatest surprise came from Maestre and colleagues, who tracked what happens when resprouting shrubs encroach on semi-arid esparto grasslands. Instead of the desertification usually expected, soil organic carbon rose from 29.5 to 33.2 milligrams per gram, total nitrogen from 1.31 to 1.50 milligrams per gram, and the potential nitrogen mineralization rate from 0.97 to 1.68 milligrams of nitrogen per kilogram per day. This occurred under the shrubs, under the esparto tussocks, and in the bare spaces between them. The shrubs involved are the mastic tree (Pistacia lentiscus), kermes oak (Quercus coccifera), rosemary, prickly juniper (Juniperus oxycedrus), and others. The finding suggests that some native shrubs create “islands of resources” that retain and recycle nutrients, reversing the course of degradation rather than accelerating it. This calls for rethinking the habit of uprooting everything that looks dense and disorderly on degraded sites. But what is happening in the depths, where the fate of groundwater is decided?
A Palm That Summons the Forest, and Roots That Search for Water
Underground, a more silent dialogue unfolds. White broom, once its roots reach deeper water resources, adapts to seasonal drought and reaches groundwater, helping sustain the ecosystems that depend on it. Canary Island tamarisk draws water from deep alluvial layers, and its presence in seasonal flow channels can indicate that the water table lies close to the surface. The resprouting shrubs in esparto grasslands send their roots horizontally for several meters, improving infiltration, reducing surface runoff, and strengthening subsurface water retention. Maestre and colleagues suggest that species such as kermes oak may practice “hydraulic lift,” moving water from deep layers to shallow roots, which might support groundwater recharge, though this mechanism remains a likelihood rather than a certainty.
In Spain and Portugal, Garrote and colleagues studied human-altered areas in the south of the Iberian Peninsula and found that the Mediterranean dwarf palm (Chamaerops humilis) is the most important nurse plant there. It was associated with 22.4 times more beneficiary species than its abundance alone would predict. The beneficiaries are late-successional woody species such as the mastic tree (Pistacia lentiscus), wild olive (Olea europaea var. sylvestris), cork oak (Quercus suber), and Phillyrea angustifolia. Its greatest value lies in the early stages of recolonization, when it triggers the nucleation processes around which the forest grows.
The lesson that unites these themes is integration: grasses to protect the topsoil from concentrated flow, deep-rooted shrubs to stabilize slopes and reach water, legumes to enrich the soil, and a nurse plant that spares the project years of waiting.
We are back at the window. When the rain falls next autumn and you watch the water cross the street, you will face a simple question: is it brown or clear? The color of the water tells the story of what happened on the hillside you cannot see. Was there esparto grass holding the earth, oleander fixing the bank, or a riparian strip that trapped what would have spoiled the river? We often build roads and neighborhoods and then ask the plants to apologize for our mistakes afterward, while the studies indicate they could have corrected many of those mistakes had they been asked at the beginning. Perhaps it is time to change the order of the questions in our offices: what was this land doing with the rain before we arrived? And do we have the courage to leave part of that work to it?
References
Nunes, A., et al. “Ecological Restoration across the Mediterranean Basin as Viewed by Practitioners.” Science of the Total Environment, 2016.
De Baets, S., et al. “Methodological Framework to Select Plant Species for Controlling Rill and Gully Erosion: Application to a Mediterranean Ecosystem.” Earth Surface Processes and Landforms, 2009.
Aparício, B. A., et al. “Modelling the Role of Ground-True Riparian Vegetation for Providing Regulating Services in a Mediterranean Watershed.” International Soil and Water Conservation Research, 2023.
Heckenroth, A., et al. “Selection of Native Plants with Phytoremediation Potential for Highly Contaminated Mediterranean Soil Restoration: Tools for a Non-Destructive and Integrative Approach.” Journal of Environmental Management, 2016.
Gairola, S. U., Bahuguna, R., and Bhatt, S. S. “Native Plant Species: A Tool for Restoration of Mined Lands.” Journal of Soil Science and Plant Nutrition, 2023.
Maestre, F. T., et al. “Shrub Encroachment Can Reverse Desertification in Semi-Arid Mediterranean Grasslands.” Ecology Letters, 2009.
Bochet, E., García-Fayos, P., and Tormo, J. “How Can We Control Erosion of Roadslopes in Semiarid Mediterranean Areas? Soil Improvement and Native Plant Establishment.” Land Degradation and Development, 2009.
Garrote, P. J., Castilla, A. R., and Fedriani, J. M. “Assessing the Relative Importance of Nurse Species on Mediterranean Human-Altered Areas.” Restoration Ecology, 2021.
