Manuscript Title
Effects of solar parks on soil quality, CO2effluxes and vegetation under Mediterranean climate
Authors
Quentin Lambert1*, Armin Bischoff2,
Sixtine Cueff3, Alexandre Cluchier4,
Raphael Gros1
Author’s institutional affiliations
1 Aix Marseille Université, Université d’Avignon, IRD, CNRS, Institut
Méditerranéen de Biodiversité et d’Écologie marine et continentale
(IMBE), Campus l’Etoile, Av. Escadrille Normandie Niémen, 13397,
Marseille, Cedex 20, France
2 Aix Marseille Université, Université d’Avignon, IRD, CNRS, Institut
Méditerranéen de Biodiversité et d’Écologie marine et continentale
(IMBE), IUT Avignon, Agroparc, BP 61207, 84911 Avignon Cedex 9, France
3 INRAE, AgroParisTech, UMR ECOSYS, 78850 Thiverval-Grignon, France
4 ECO-MED Ecologie & Médiation, Tour Méditerranée, 65 avenue Jules
Cantini, 13298 Marseille cedex 20, France
Corresponding author:qntnlmbrt@gmail.com
Short informative containing containing the major key words.
Effects of solar parks on soil quality, CO2 effluxes and vegetation
under Mediterranean climate
Solar parks are expanding in Europe, but their impact on soil and
vegetation is not well studied yet. We have shown in this study, carried
out in 3 parks in the Mediterranean region, that the construction of
solar parks reduces the physical quality of the soil that could alter
main soil function. Moreover, the presence of solar panels decreases CO2
emissions and temperature but does not change the structure of plants
communities.
Short running title
Effects of solar parks on soil quality, CO2 effluxes and vegetation
Acknowledgements
Q.L acknowledges the French Agency for Environmental Transition (ADEME)
for his PhD ground. This work was supported by the program PIESO (ADEME,
agreement N°1405C0035). The authors which to thanks Pierre Illac and
Marine David (Total Quadran) for their contribution to this project.
Authors thanks Mathilde Dionisi, Jean Bigotte, Xavier Fortuny, Lisa Foli
for their very useful technical assistance in the lab or in the field.
Abstract and Keywords
Abstract:
Solar energy is increasingly used to produce electricity in Europe, but
the environmental impact of constructing and running solar parks (SP) is
not yet well studied. Solar park construction requires partial
vegetation removal and soil leveling. Additionally, solar panels may
alter soil microclimate and functioning. In our study of three French
Mediterranean solar parks, we analyzed 1) effects of solar park
construction on soil quality by comparing solar park soils with those of
semi-natural land cover types (pinewood and shrubland) and abandoned
croplands (abandoned vineyards); 2) the effect of solar panels on soil
microclimate, CO2 effluxes and vegetation. We measured
21 soil properties of physical, chemical, and microbiological soil
quality in one solar park and its surroundings to calculate integrated
indicators of soil quality. We surveyed soil temperature and moisture,
CO2 effluxes and vegetation below and outside solar
panels of three solar parks. Soil aggregate stability was reduced by SP
construction resulting in a degradation of soil physical quality. Soil
chemical quality and a general indicator of soil quality were lower in
anthropogenic (SP and abandoned vineyards) than in semi-natural
(pinewood and shrubland) land cover types. However, differences between
abandoned vineyards representing the pre-construction land cover type
and solar parks were not significant. Solar panels reduced the soil
temperature by 10% and soil CO2 effluxes by 50% but
did not affect early successional plant communities. Long-term
monitoring is needed to evaluate the effects of solar panels on
vegetation.
Keywords: renewable energy, soil functions, land cover, microclimate,
soil respiration, plant communities
Main text
Introduction
The use of solar energy to produce electricity is increasingly common in
Europe and requires large areas in order to be cost-effective (Murphy et
al., 2015 ; Ong et al., 2013). Solar park construction involves clearing
and grading the soil surface, burying of electric cables, vegetation
removal and soil compaction increasing runoff and erosion. Grading,
compaction, and erosion change the physical and chemical properties of
the soil and thus reduce its quality. Since solar park construction
destroys the vegetation and affects the soil, a careful analysis of the
environmental impact of solar parks is needed (Armstrong et al. ,
2016; Hernandez et al. , 2015). To our knowledge, analyses of soil
quality have not yet been included in studies on the impact of solar
park construction although soil quality is an important indicator of
ecosystem functioning. After the installation of solar panels, the
vegetation is regularly mown or grazed limiting vegetation height to
prevent shading of panels. The solar panels also change the microclimate
such as temperature, humidity, solar radiation (Tanner et al. ,
2020; Armstrong et al. , 2016). Such changes in microclimate may
affect soil processes and plant communities under panels, in particular
in the European Mediterranean with high solar irradiation compared to
temperate regions.
Soil quality is “the capacity of a specific kind of soil to function,
within natural or managed ecosystem boundaries, to sustain plant and
animal productivity, maintain or enhance water and air quality and
support human health and habitation” (Karlen et al. , 1997).
Three soil quality indicator groups are commonly used: physical,
chemical and biological soil properties (Bünemann et al., 2018 ;
Costantini et al., 2016 ; Maurya et al., 2020). Physical properties,
such as bulk density and texture influence water holding capacity and
plant communities by modulating root growth (Scarpare et al. 2019 ;
Lampurlanés, Cantero-Martínez 2003). Chemical properties such as
inorganic N, total C and pH control plant nutrition and microbiological
activity. Biological indicators include the activity of decomposers such
as invertebrates or microorganisms. These organisms control organic
matter decomposition and nutrient cycling (Maurya et al. , 2020).
(Velasquez et al. , 2007) developed a single general indicator of
soil quality (GISQ) that integrates a set of physical, chemical and
biological soil properties. Such soil properties are chosen and measured
to evaluate multifaceted aspects of soil functions and further combined
to calculate sub indicators of physical quality, chemical fertility, and
biological functioning. The GISQ combines the sub indicators to provide
a global assessment of soil quality based on soil ecosystem services and
facilitates the comparison of soils between different sites/habitats. In
a comparative study on four land use types, (Raiesi & Salek‐Gilani,
2020) showed, using an adapted GISQ, that soil quality was 1.5 times
lower in anthropogenic than in natural soils. Joimel et al. (2016)
observed a decrease in soil physico-chemical quality along an
anthropization gradient from forest to urban soils whereas Joimel et al.
(2017) did not find any difference in biological quality of these soils.
The construction of solar parks on natural and semi-natural land use
types (e.g. forest, shrubland, abandoned vineyards) may reduce soil
quality and affect ecosystem functions such as infiltration and storage
of water, fertility and plant reestablishment, soil organic matter and
nutrient cycling (Khare & Goyal, 2013; Romero-Díaz et al. , 2017;
Rutgers et al. , 2009; Scarpare et al. , 2019; Yin et
al. , 2020).
Plant communities and soil functioning may also be affected by changes
in microclimate under solar panels. Solar panels reduce solar radiation,
air humidity and soil temperatures, but in winter, soil temperatures are
generally higher under panels (Armstrong et al., 2016). Adeh et al.
(2018) reported highest soil moisture and local heterogeneity of soil
water conditions under solar panels. Such changes in microclimate may
alter plant community composition and soil respiration that can be
measured as CO2release. Mediterranean plant
communities are dominated by heliophilous plants (Bagella & Caria,
2012). The reduction of solar radiation under solar panels may thus
result in a plant community shift towards shade-tolerant species.
Seed germination of Mediterranean
species may be limited by light reduction (Gresta et al. , 2010)
and the mortality of heliophilous
plants increases in competition to shade-tolerant species. (Novaraet al. , 2012; de Dato et al. , 2010). The change in air and
soil microclimate under panels reduced the soil respiration under
temperate oceanic climate (Armstrong et al. , 2016).
Under Mediterranean climate with
higher annual temperatures and summer drought, changes in microclimate
under solar panels may be higher resulting in a strong disturbance of
seasonal soil respiration dynamics (González-Ubierna & Lai, 2019).
Plant communities also contribute to soil CO2 release by
respiration of roots and rhizosphere microorganisms (Raich &
Tufekciogul, 2000) but also by changes in soil structure (Yang et
al. , 2009; Zou et al. , 2005). Furthermore, plants are the
principal carbon source of decomposer microorganisms (Wall et
al. , 2012). Thus, solar panels may also change soil conditions
indirectly by a shift in plant community composition since plants are
very sensitive to change in microclimate.
The aims of our study were to assess 1) the effect of solar park
construction on soil quality in comparing solar parks with semi-natural
land cover types (pinewood and shrubland) and abandoned cropland (i.e.
abandoned vineyards) and 2) the effects of solar panels on soil
microclimate, CO2 effluxes and vegetation under Mediterranean climate.
We expected that 1) solar park construction reduces physical, chemical,
and biological soil quality, 2) solar panels change soil microclimate
and plant community composition, and 3) solar panels change soil
respiration according to the season.
2. Materials and Methods
2.1. Study sites
Two studies were conducted in three solar parks (SP) located in Southern
France (La Calade, Pouzols-Minervois and Roquefort des Corbières) with a
distance of 10 to 30 km from one another (Table 1). These SP were
constructed in 2011, 2014 and 2016, respectively, covered between 8.5
and 16 Ha and used ground-fixed photovoltaic (PV) systems carrying the
solar panels at a fixed inclination. The solar panels are aligned to
form rows (height of 0.6 m min and 2 m max) exposed to the South and
with a gap of 4 m between rows. The study region is characterized by
typical Mediterranean climate with summer drought and mild, wet winters.
The SP are mainly bordered by pine forests (Pinus halepensis ),
shrublands and vineyards. Dominant species of these shrublands areQuercus coccifera, Pistacia lentiscus, Rosmarinus officinalis,
Myrtus communis, Genista scorpius, Brachypodium retusum andCistus monspelliensis . The soils of the SPs are characterized by
carbonatic pedofeatures (i.e. fine calcareous silty clay soil).
2.2. Sampling designs
Study 1: effect of solar park construction
To study the effects of solar park construction on soil quality, four
sampling plots (50 ×2m) separated by 100 m were randomly chosen within
the SP at Roquefort des Corbières (inter-rows between solar panels).
Close to this SP, three major land cover types were identified
(pinewood, shrubland and abandoned vineyards) and four sampling plots of
the same area (100m²) separated by 400m were randomly chosen for each
land cover. In March 2016, ten soil samples were randomly collected
(10 cm depth) within each plot, mixed to one composite sample per plot.
Composite samples were sieved (mesh size: 2 mm) prior to analyses. An
aliquot of samples was air-dried (1 week, 30 °C). For each sample,
another aliquot was stored at 4 °C for microbial analyses.
Study 2: effect of solar panels
To study the effect of solar panels on soil respiration, temperature,
and moisture and on plant communities, we randomly selected within each
of the 3 SP four sampling plots (50 ×2m) below the solar panels, both
separated by at least 100m, and four adjacent sampling plots (50 ×2m) in
the inter-rows between the solar panel.
2.3. Measurements of soil physico-chemical and microbiological quality
Soil physical properties
Water content (g.kg-1) was determined after drying
samples (24 hours, 105°C). Water holding capacity (WHC) was analyzed
according to Saetre (1998) but using a modified protocol. 10g of dried
soil were weighted in a PVC cylinder and saturated with water. WHC was
defined as the water content remaining in the soil after 12h (4°C). The
different soil fractions (i.e. sand, silt, clay) were determined
using the Robinson’s pipette method (Olmstead et al. , 1930) after
organic matter removal by oxidation with
H202 (30%, 48 hours). Bulk density (BD)
was determined by measuring dried soil mass sampled in a Siegrist’s
cylinder. According to Huang et al., (2004), a value of
2.65g.cm-3 was assumed for real soil density (RD).
Soil porosity was calculated using the following equation.
\(\text{Soil\ porosity\ }=100\times\frac{RD-BD}{\text{RD}}\)(Equation. 1)
Mean weight diameter (MWD) of soil aggregates was measured according to
Kemper and Rosenau (1986).
Soil chemical properties
The soil pH was measured in distilled water and KCL (1M) (Aubert, 1978).
Total Carbon (TC) and Nitrogen (TN) content were determined by
combustion in an elemental analyzer CN FlashEA 1112 (ThermoFisher) (NF
ISO 10694, NFISO 13878). Calcium carbonate (CaCO3) content was measured
using a Bernard calcimeter (Müller & Gastner, 1971) and the percentage
of C in CaCO3 (C-CaCO3) was determined as: C-CaCO3 = 11.991 / 100 x
CaCO3. Inorganic nitrogen (NH4+ and
NO3-) was extracted in KCL solution
(1M) and analyzed calorimetrically using the nitroprusside-salicylate
and nitrosalicylic acid method according to Mulvaney (1996) and Keeney
and Nelson (1983), respectively.
Soil microbiological properties
Microbial Biomass (MB) was measured using substrate induced respiration
(SIR) rates (Anderson and Domsch, 1978). Basal respiration was
determined without adding glucose and was estimated to calculate the
metabolic quotient qCO2 (the ratio of basal respiration
to microbial biomass), which is a sensitive ecophysiological indicator
of soil stress (Anderson, 2003). Three enzyme activities (i.e.fluorescein diacetate hydrolase, phosphatase and tyrosinase) involved in
carbon and phosphorous cycles were assessed (n=3 per sample) to
determine the catabolic potential of microbial communities.
Fluorescein diacetate hydrolase
(FDase, U.g-1 dry weight) was measured according to
Green et al. (2006), phosphatase (U.g-1 dry weight)
according to Tabatabai and Bremner (1969) and the activity of tyrosinase
(µmol.min-1.g-1 dry weight)
according to Saiya-Cork et al. (2002).
2.4. Measurements of solar panel effects on soil moisture, temperature
and in situ respiration
Soil respiration, temperature and moisture were recorded in March and
June 2017 in each sampling plot of the study on solar panel effects.In situ CO2 release (g CO2m-1 h-1) from soils, plants roots,
soil organisms and chemical oxidation of C compounds was measured after
removal of aboveground vegetation, using a portable gas analyser (IRGA,
EGM-4, PP-system). The device was connected to a closed soil respiration
chamber (SRC-1, PP systems Massachusetts, USA). To prevent leakage of
CO2 when placing the chamber on the soil, a PVC tube (10
cm x 11 cm) was buried 1 cm deep into the soil prior to measurements.
Soil temperature was recorded in a depth of 7cm using the soil
temperature probe (STP-1, PP-system) connected to the respirometer. Soil
moisture was recorded on four points at 7cm depth using a portable
time-domain reflectometry (TDR) device (Delta-T Devices, ML2 Theta
Probes).
2.5. Measurements of solar panel effects on vegetation
In the sampling plots of the study on solar panel effects, vegetation
surveys were carried out in 2016 and 2017. Three rectangular sub-plots
of 10m² (2m × 5m) were placed at the ends and the center of each plot.
Percentage cover of all occurring vascular plant species was estimated
as the vertical projection of aboveground plant organs. A ratio of
shadow-tolerant (sciaphilous) to hemi-heliophilous and heliophilous
plant species (Julve, 2020) was calculated .
2.6. Statistical analyses
We calculated a General Indicator of Soil Quality (GISQ) according to
Velaquez et al. (2007). Information on 21 variables of physical,
chemical, and microbiological soil properties was used to create three
sub-indicators related to main soil functions: 1) physical properties
that determine the infiltration and storage of water, 2) chemical
properties that affect fertility and plant reestablishment in solar
parks, 3) microbiological properties that drive soil organic matter
decomposition and nutrient cycling. For each group of variables
(physical, chemical and microbiological), a principal component analysis
(PCA) with data scaled to unit of variance was run using “FactoMineR”
(Husson et al. , 2020) and “Factoextra” (Kassambara & Mundt,
2020) packages. A synthetic index of quality for each group of variables
at a plot i (Iqi) was calculated as the sum of n
variables (vi) multiplied by their respective weight (wi) in the
determination of axes 1 and 2 of the PCA (Equation 2.).
\(\text{Iq}_{i}\ =\sum_{i=1}^{n}{w_{i}v_{i}}\) (Equation2.)
The values of Iqi were then reduced to a common range,
between 0.1 and 1.0, using an homothetic transformation to obtain the
sub-indicators of physical, chemical and microbiological soil quality
(hereafter pSQ, cSQ and mSQ respectively, Equation 3.). In this
equation, a is the maximal and b the minimal Iq value for the plot i.
\(p,\ c\ or\ mSQ=0.1+\ \frac{\text{Iq}_{i}-b}{\text{Iq}_{i}-a}\ \times 0.9\)(Equation 3.)
Finally, a PCA was run with the 3 sub-indicators. The GISQ was obtained
by summing the products of the respective contributions of variables to
factors 1 and 2 by the % inertia explained by factors, respectively.
Finally, the sum of these products gave the following formula for the
GISQ (Equation 4.):
\(GISQ=0.29\ pSIq+0.28\ cSIq+0.33\ mSIq\) (Equation 4.)
All data were analyzed using R software (3.6.1,R Core (Team, 2020)).
Effects of land cover type on soil physical, chemical and
microbiological properties, sub-indicators of soil quality and GISQ were
assessed using one way-analysis of variance (ANOVA). In the case of a
significant land cover type effect, a Tukey HSD post hoc test was used
to analyze differences between specific land cover types. To analyze the
effect of solar panels on soil temperature, water content, CO2 effluxes
and vegetation, linear mixed-effect models (LMMs) (R package “lme4”)
were applied including month and treatment (below vs outside panels) as
fixed factors and solar park identity as random factor. When necessary,
data were transformed using the “bestNormalize” package (Peterson,
2019) to meet the assumptions of normality and homoscedasticity of
variances. Effects of solar panels on plant communities were visualized
by non-metric multidimensional scaling (nMDS) based on the Bray-Curtis
dissimilarity. Differences in plant community composition were tested
using permutational multivariate analysis (PERMANOVA) in R package
“vegan” (Oksanen et al. , 2019).
3. Results
3.1. Effects of solar park construction on soil properties
Seven of the eight tested physical soil properties were significantly
different between land cover types (Table 2). Among these seven
properties, only two showed a significant difference between the two
semi-natural (pinewood and shrubland) land cover types and the SP. Soil
water content was 5.5% lower in the SP (p <0.01) than in
shrubland. The mean weight diameter of aggregates was twice as high in
abandoned vineyard as in SP, and three times higher in pinewood and
shrubland than in SP (p<0.001). Organic carbon was about 2.5
times higher in semi-natural than in anthropogenic land cover types
(p<0.001). Sand and silt content, soil porosity and bulk
density were significantly different between abandoned vineyard and
pinewood (Table 2). Silt content and soil porosity were 1.4 and 1.3
times lower in abandoned vineyard than in pinewood, respectively. Sand
content and BD were about 1.5 times higher in abandoned vineyard than in
pinewood. Pinewood and shrubland showed similar physical properties
without significant differences.
For most soil chemical properties, SP showed significant differences to
pinewood and shrubland but not to abandoned vineyard (Table 2). Total
carbon contents were 1,44 times higher in semi-natural land cover types
than in antropogenic land cover types(p<0.01). Organic carbon
contents were about 2,76 times higher in semi-natural land cover types
than in antropogenic land cover types (p<0.01). Total nitrogen
(TN) content showed the same pattern. TN was twice as high in pinewood
and shrubland as in the SP and abandoned vineyard (p< 0.001).
The water pH ranged between 8.02 and 8.11 and showed a small but
significant difference between shrubland and abandoned vineyard.
Two microbiological properties were significantly different between land
cover types (Table 2). Land cover type had a significant influence on
basal respiration (p<0.03) being two times lower in the SP and
abandoned vineyard than in forest and shrubland. The FDAse activity was
two times higher in shrubland and pinewood than in the SP and abandoned
vineyard. Microbial biomass was twice as low (marginally significant) in
SP and abandoned vineyard as in the semi-natural land cover types.
3.2. Effects of solar park construction on soil quality
The first two axes of the PCA run on physical properties explained 69.94
% of the total variance (see A.1.A). The semi-natural land cover types
are separated from the antropogenic soils along the first axis. Silt,
water content, water holding capacity and mean weight diameter of
aggregates had the highest score on the first PCA axis, while soil
porosity was strongly correlated with the second axis (see A.1.A). The
highest physical quality index of 0.92 was measured in pinewood and
shrubland being two times higher than in abandoned vineyard
(p<0.001). The pSQ (Figure 1A) was two times and four times
lower in SP than in the abandoned vineyard and semi-natural land cover
types, respectively (p <0.01).
The first two axes of the PCA used to calculate the sub-indicator of
soil chemical quality (cSQ) explained 73.78 % of the total variance
(see A.1.B). The semi-natural land cover types are separated from the
antropogenic soils along the first axis. Total carbon, organic carbon,
total nitrogen and ammonium were most correlated with the first axis and
nitrate with the second axis (see A.1.B). With a value of 0.18, the cSQ
was four times lower in the SP and abandoned vineyard than in shrubland
and pinewood (p<0.001, Figure 1B).
The first two axes of the PCA used to calculate the sub-indicator of
soil microbiological quality explained 77.19% of the total variance
(see A.1.C). Basal respiration, microbial biomass, FDAse and phosphatase
were highly correlated with the first PCA axis, while the
qCO2 was correlated with the second one (see A.1.C). The
mSQ was not significantly different between land cover types (p = 0.95)
(Figure 1 C).
The General Indicator of Soil Quality was four times lower in the SP and
abandoned vineyard than in the pinewood and shrubland
(p<0.001).
3.3. Effects of solar panels on soil temperature, water content andin situ CO2 effluxes.
Soil temperature and water content were significantly different between
months (p<0.05; Figure 2A and 2B). Solar panels significantly
decreased soil temperature in March and June (Figure 2A) but did not
affect soil water content (p = 0.79). Soil CO2 effluxes
did not change between months but were twice as high outside solar
panels than below solar panels (p < 0.001).
3.3. Effects of solar panels on plant communities
Neither the species richness nor the total cover of plant community was
significantly affected by the solar panels (Table 4). A marginally
significant difference was detected for the ratio ‘Sciaphile: Heliophile
plants’, being higher below than outside solar panels. The NMDS and
PEMANOVA did not reveal any significant panel effect on plant community
composition (p = 0.3461, Figure 3). However, community composition was
significantly different between the solar parks (p< 0.001). No
significant difference was detected between observation years (data not
shown).
4. Discussion
Solar park (SP) construction reduced physical and chemical soil quality
compared with semi-natural land cover types (forest and shrubland) but
not biological soil quality. A change in soil temperature and
CO2 effluxes also demonstrated a negative solar panel
effect on soil microclimate and functioning. However, in early stages of
plant succession following solar park construction, plant community
composition below and outside solar panels was not significantly
different.
4.1 Effects of solar park construction on soil quality
Soil quality assessments require the measurement of a wide range of
physical, chemical, and biological properties involving a high
complexity of potential analyses (Maurya et al. , 2020). In this
study, we assessed soil quality using a multi-proxy approach including
21 soil properties. The reduction of the number of variables using PCA
to group these properties allows an integrated evaluation of soil
quality based on their main functions, such as infiltration and storage
of water, soil fertility, plant reestablishment and soil organic matter
and nutrient cycling. We found that two of three integrated
sub-indicators and the general indicator of soil quality were lower in
SP than in the other land cover types.
Among the physical soil properties, the aggregate MWD was 1.5 times
lower in the SP than in the semi-natural land cover types. A low MWD may
result in a low aggregate stability. Similarly, (Kabir et al. ,
2017) showed that the MWD decrease in anthropogenic soils associated
with a degraded vegetation. In our study, soil levelling and vegetation
removal prior to SP construction may have decreased soil organic matter
(SOM) content reducing MWD. By binding colloids and stabilizing soil
structure, SOM plays a key role in soil physical properties and nutrient
cycling (Six et al. , 2004). Telak and Bogunovic (2020) showed a
decrease in SOM and MWD in a vineyard of Croatia after intensive and
frequent tillage. Such mechanical disturbance for many years may have
affected soil structure of the vineyard on which the studied SP
(Roquefort des Corbières) was constructed. A lower SOM affects microbial
activity and production of mucus resulting in a decrease of aggregates
MWD and thus a soil more sensitive to erosion (Blavet et al. ,
2009; Le Bissonnais et al. , 2018). The soil levelling and
vegetation removal during SP construction may have increased surface
runoff and soil erosion (Rabaia et al. , 2021). In our study, the
effect of SP construction was not strong enough to change these physical
soil properties. In contrast to our expectations, the SP construction
did neither increase soil compaction nor decrease porosity compared to
the abandoned vineyard. The past soil tillage in abandoned vineyard may
have already degraded these properties before SP construction limiting
effects of construction work.
Accordingly, overall physical soil quality of SP was lower than that of
abandoned vineyard which was in turn lower than that of semi-natural
land cover types. The physical soil quality index revealed that the
construction of a SP increased the degradation of the physical soil
quality of soils already degraded by land management (abandoned
vineyard). In particular, the stability of the soil, key factor of soil
functioning, was lower in SP than in abandoned vineyard. Soil
restoration by revegetation may improve soil physical quality and
functions of solar parks over initial vineyard conditions towards
semi-natural land cover types (Hernandez et al. , 2019).
Soil chemical properties, such as total and organic carbon and total
nitrogen are directly linked to soil fertility and plant growth (Krullet al. , 2004; Liu et al. , 2014). In our study, these
properties showed lower values in anthropogenic soils than in
semi-natural land cover types. Joimel et al. (2016) obtained similar
results along a gradient from natural to anthropogenic habitats in which
total carbon and nitrogen decreased significantly from forests to
vineyard. Soil disturbance such as soil tillage in vineyard or
construction activities increases mineralization of organic matter
reducing organic C and N (Brantley & Young, 2010). Accordingly, Choi et
al. (2020) found a significantly lower C and N content in SP than in
grassland soil. In our study, SP construction did not reduce neither C
and N content nor soil chemical quality compared to degraded vineyard
soil. Our results suggest that previous agricultural practices have
strongly affected the soil chemical quality and that the construction of
SP did not have an additional impact.
Soil microorganisms (i.e. bacteria and fungi) contribute actively
to soil nutrient cycling (Schimel and Schaeffer, 2012). Thus, their
genetic and physiologic characteristics are important indicators of
ecosystem functioning such as nutrient cycling (Ranjard et al. ,
2011). Microbiological soil properties showed differences between land
cover types for fluorescein diacetate hydrolase (FDAse) activity and
basal respiration. FDAse is an appropriate proxy to evaluate soil
microbial activities because the ubiquitous esterase enzymes (lipase,
protease, phosphatase) are involved in the hydrolysis of FDA (Schnürer
& Rosswall, 1982). In our study, the FDAse was two times lower in
anthropogenic soils suggesting a reduction of microbiological activity
and nutrient cycling. Soil basal respiration showed the same pattern
confirming a degradation of soil functions compared to semi-natural
soils (Sparling 1997). A lower rate of basal respiration may be the
result of a lower organic carbon and nitrogen content (Horakova et al.
2020). Despite the reduction of these two microbial properties in
anthropogenic soils, the microbiological soil quality index (mSiQ) was
not significantly different between land cover types. Other
microbiological properties (BM and phosphatase) mainly contributing to
the first PCA axis were not affected by land cover type and thus
overruled significant response variables in mSiQ calculation.
As a consequence of lower physical and chemical sub-indicators, the
general indicator of soil quality was about three times lower in SP
compared to semi-natural land cover types. Similarly, Zhang et al.
(2019) found that the soil quality was about 50% higher in restored
shrubland than in anthropogenic soils (cropland). The key processes
involved in degradation of soil quality were soil tillage, partial
topsoil removal increasing erosion (Quinton et al. , 2010) and
organic matter mineralization. Reduced organic matter content and
increase of soil compaction decrease water holding capacity (Mujdeciet al. , 2017) and soil stability (Simansky et al. , 2013).
4.2 Effects of solar panels on vegetation, soil microclimate and
CO2 effluxes
Climatic conditions influence both soil microbial activities (Shaoet al. , 2018) and plant communities (García‐Fayos & Bochet,
2009). In our study, solar panels reduced the soil temperature in spring
and in summer by about 5°C. Similarly, Armstrong et al. (2016) found a
soil temperature reduction of 2°C under solar panels during the summer
(UK). The lower temperature under solar panels was the direct effect of
shading although night temperatures may be higher (Tanner et al. ,
2020). Solar panels also intercept precipitation, and Tanner et al.
(2020) found a significant reduction in soil humidity under solar panels
in the Mojave desert. However, we did not find any significant soil
humidity difference under solar panels and outside. The result may be
explained by a lower evapotranspiration limiting humidity losses during
drought periods as suggested by Tanner et al.(2020).
Mediterranean vegetation is dominated by heliophilous plants (Bagella &
Caria, 2012). So, we expected that light reduction by solar panels
strongly affects plant communities. However, we did not find a
significant effect of solar panels on plant community composition and
structure. The effect of solar panels on the ratio of shadow-tolerant to
heliophilous species was only marginally significant and no influence on
plant species richness was detected. Other studies showed, however, a
reduction in plant cover and species richness under solar panels
resulting from lower germination and higher mortality (Armstrong et al.
2016). Protection against strong solar radiation and drought during
Mediterranean summer may have compensated for reduction of light and
precipitation in our study. Accordingly, Tanner et al. (2020) observed
that in a desert plant richness was marginally greater under their solar
panels than in the control. In our study, the absence of a solar panel
effect on the vegetation may also be explained by the low age of our
solar parks limiting differential effects on the vegetation. In early
successional stages, the vegetation is dominated by ubiquitous annual
species germinating and developing under a great variety of
environmental conditions. Responses to the specific microclimate under
solar panels may be slow in Mediterranean vegetation types
(Coiffait-Gombault et al. , 2012; Kinzig et al. , 1999).
Long-term monitoring is required to finally evaluate the influence of
solar panels on plant communities.
Soil CO2 effluxes are driven by soil climate (Francioniet al. , 2020) and vegetation (Moinet et al. , 2019). In our
study, soil respiration was highly affected by solar panels. Similarly,
Armstrong et al. (2016) found a reduction of soil CO2effluxes under solar panels . We detected a reduced CO2efflux already in March. Since temperature is the major driver of soil
respiration (González-Ubierna & Lai, 2019) the difference is probably
the result of the warmer Mediterranean spring increasing solar panel
effects. However, the reduction of CO2 effluxes under
solar panels may also be the result of light reduction reducing plant
growth and root respiration. A lower soil respiration is an indicator of
lower litter decomposition and nutrient cycling suggesting that these
ecosystem functions may be reduced under solar panels (Incerti et
al. , 2011).
4.4 Conclusions
Physical, chemical, and global soil qualities were lower in solar park
than in semi-natural land cover types. Clearing and grading the soil
surface during solar park construction induced a strong degradation of
soil physical quality, especially of soil structure, but did not disturb
nor soil chemical quality neither global quality. Our study suggests
that the solar parks should be constructed preferably on anthropogenic
soils or that it must be accompanied by environmental reduction measures
and ecological restoration. At our Mediterranean study sites, solar
panels reduced both soil temperature and soil CO2effluxes but not vegetation in the beginning of plant succession. These
effects could, however, alter soil functions such as organic matter
decomposition and nutrient cycles leading to disturb plant establishment
and growth in the long term. Long-term monitoring including different
seasons is required to evaluate the final response of soil properties
and vegetation to solar panels.
Reference
Adeh EH, Selker JS, Higgins CW. 2018. Remarkable agrivoltaic influence
on soil moisture, micrometeorology and water-use efficiency. Plos
One 13 : e0203256. DOI: 10.1371/journal.pone.0203256
Anderson JP, Domsch KH. 1978. A physiological method for the
quantitative measurement of microbial biomass in soils. Soil
biology and biochemistry 10 : 215–221
Anderson T-H. 2003. Microbial eco-physiological indicators to asses soil
quality. Agriculture, Ecosystems & Environment 98 :
285–293. DOI: 10.1016/S0167-8809(03)00088-4
Armstrong A, Ostle NJ, Whitaker J. 2016. Solar park microclimate and
vegetation management effects on grassland carbon cycling.Environmental Research Letters 11 : 074016. DOI:
10.1088/1748-9326/11/7/074016
Aubert G. 1978. Methodes d’analyses des sols: documents de travail
tous droits reserves . Centre régional de documentation pédagogique
Bagella S, Caria MC. 2012. Diversity and ecological characteristics of
vascular flora in Mediterranean temporary pools. Comptes Rendus
Biologies 335 : 69–76. DOI: 10.1016/j.crvi.2011.10.005
Blavet D, De Noni G, Le Bissonnais Y, Leonard M, Maillo L, Laurent JY,
Asseline J, Leprun JC, Arshad MA, Roose E. 2009. Effect of land use and
management on the early stages of soil water erosion in French
Mediterranean vineyards. Soil and Tillage Research 106 :
124–136. DOI: 10.1016/j.still.2009.04.010
Brantley ST, Young DR. 2010. Shrub expansion stimulates soil C and N
storage along a coastal soil chronosequence. Global Change
Biology 16 : 2052–2061. DOI: 10.1111/j.1365-2486.2009.02129.x
Bünemann EK, Bongiorno G, Bai Z, Creamer RE, De Deyn G, de Goede R,
Fleskens L, Geissen V, Kuyper TW, Mäder P, Pulleman M, Sukkel W, van
Groenigen JW, Brussaard L. 2018. Soil quality – A critical review.Soil Biology and Biochemistry 120 : 105–125. DOI:
10.1016/j.soilbio.2018.01.030
Choi CS, Cagle AE, Macknick J, Bloom DE, Caplan JS, Ravi S. 2020.
Effects of Revegetation on Soil Physical and Chemical Properties in
Solar Photovoltaic Infrastructure. Frontiers in Environmental
Science 8 . DOI: 10.3389/fenvs.2020.00140
Coiffait-Gombault C, Buisson E, Dutoit T. 2012. Are old Mediterranean
grasslands resilient to human disturbances? Acta Oecologica43 : 86–94. DOI: 10.1016/j.actao.2012.04.011
Costantini EAC, Branquinho C, Nunes A, Schwilch G, Stavi I, Valdecantos
A, Zucca C. 2016. Soil indicators to assess the effectiveness of
restoration strategies in dryland ecosystems. Solid Earth7 : 397–414. DOI: https://doi.org/10.5194/se-7-397-2016
de Dato GD, De Angelis P, Sirca C, Beier C. 2010. Impact of drought and
increasing temperatures on soil CO2 emissions in a Mediterranean
shrubland (gariga). Plant and Soil 327 : 153–166. DOI:
10.1007/s11104-009-0041-y
Francioni M, Trozzo L, Toderi M, Baldoni N, Allegrezza M, Tesei G,
Kishimoto-Mo AW, Foresi L, Santilocchi R, D’Ottavio P. 2020. Soil
Respiration Dynamics in Bromus erectus-Dominated Grasslands under
Different Management Intensities. Agriculture 10 : 9.
DOI: 10.3390/agriculture10010009
García‐Fayos P, Bochet E. 2009. Indication of antagonistic interaction
between climate change and erosion on plant species richness and soil
properties in semiarid Mediterranean ecosystems. Global Change
Biology 15 : 306–318. DOI:
https://doi.org/10.1111/j.1365-2486.2008.01738.x
González-Ubierna S, Lai R. 2019. Modelling the effects of climate
factors on soil respiration across Mediterranean ecosystems.Journal of Arid Environments 165 : 46–54. DOI:
10.1016/j.jaridenv.2019.02.008
Green VS, Stott DE, Diack M. 2006. Assay for fluorescein diacetate
hydrolytic activity: Optimization for soil samples. Soil Biology
and Biochemistry 38 : 693–701. DOI:
10.1016/j.soilbio.2005.06.020
Gresta F, Cristaudo A, Onofri A, Restuccia A, Avola G. 2010. Germination
response of four pasture species to temperature, light, and post-harvest
period. Plant Biosystems 144 : 849–856. DOI:
10.1080/11263504.2010.523549
Hernandez RR, Armstrong A, Burney J, Ryan G, Moore-O’Leary K, Diédhiou
I, Grodsky SM, Saul-Gershenz L, Davis R, Macknick J, Mulvaney D, Heath
GA, Easter SB, Hoffacker MK, Allen MF, Kammen DM. 2019.
Techno–ecological synergies of solar energy for global sustainability.Nature Sustainability 2 : 560–568. DOI:
10.1038/s41893-019-0309-z
Hernandez RR, Hoffacker MK, Murphy-Mariscal ML, Wu GC, Allen MF. 2015.
Solar energy development impacts on land cover change and protected
areas. Proceedings of the National Academy of Sciences112 : 13579–13584. DOI: 10.1073/pnas.1517656112
Horakova E, Pospisilova L, Vlcek V, Mensik L. 2020. Changes in the
soil’s biological and chemical properties due to the land use.Soil and Water Research 15 : 228–236. DOI:
10.17221/44/2019-SWR
Huang Q, Akinremi OO, Rajan RS, Bullock P. 2004. Laboratory and field
evaluation of five soil water sensors. Canadian Journal of Soil
Science . DOI: 10.4141/S03-097
Husson F, Josse J, Le S, Mazet J. 2020. FactoMineR: Multivariate
Exploratory Data Analysis and Data Mining .
Incerti G, Bonanomi G, Giannino F, Rutigliano FA, Piermatteo D, Castaldi
S, De Marco A, Fierro A, Fioretto A, Maggi O, Papa S, Persiani AM, Feoli
E, De Santo AV, Mazzoleni S. 2011. Litter decomposition in Mediterranean
ecosystems: Modelling the controlling role of climatic conditions and
litter quality. Applied Soil Ecology 49 : 148–157. DOI:
10.1016/j.apsoil.2011.06.004
Joimel S, Cortet J, Jolivet CC, Saby NPA, Chenot ED, Branchu P, Consalès
JN, Lefort C, Morel JL, Schwartz C. 2016. han. Science of The
Total Environment 545–546 : 40–47. DOI:
10.1016/j.scitotenv.2015.12.035
Joimel S, Schwartz C, Hedde M, Kiyota S, Krogh PH, Nahmani J, Pérès G,
Vergnes A, Cortet J. 2017. Urban and industrial land uses have a higher
soil biological quality than expected from physicochemical quality.Science of The Total Environment 584–585 : 614–621.
DOI: 10.1016/j.scitotenv.2017.01.086
Julve P. 2020. Baseflor. Index botanique, écologique et chorologique de
la flore de France.
Kabir EB, Bashari H, Mosaddeghi MR, Bassiri M. 2017. Soil aggregate
stability and organic matter as affected by land-use change in central
Iran. Archives of Agronomy and Soil Science 63 :
1823–1837. DOI: 10.1080/03650340.2017.1308492
Karlen DL, Mausbach MJ, Doran JW, Cline RG, Harris RF, Schuman GE. 1997.
Soil Quality: A Concept, Definition, and Framework for Evaluation (A
Guest Editorial). Soil Science Society of America Journal61 : 4–10. DOI:
https://doi.org/10.2136/sssaj1997.03615995006100010001x
Kassambara A, Mundt F. 2020. factoextra: Extract and Visualize the
Results of Multivariate Data Analyses .
Keeney D, Nelson D. 1983. Nitrogen –inorganic forms. Methods of
Soil Analysis. Part 2. Chemical and Microbiological Properties. John
Wiley & Sons, Ltd, i–xxiv. DOI: 10.2134/agronmonogr9.2.2ed.frontmatter
Kemper WD, Rosenau RC. 2018. Aggregate Stability and Size Distribution.Methods of Soil Analysis . John Wiley & Sons, Ltd, 425–442. DOI:
10.2136/sssabookser5.1.2ed.c17
Khare P, Goyal DK. 2013. Effect of high and low rank char on soil
quality and carbon sequestration. Ecological Engineering52 : 161–166. DOI: 10.1016/j.ecoleng.2012.12.101
Kinzig AP, Levin SA, Dushoff J, Pacala S. 1999. Limiting Similarity,
Species Packing, and System Stability for Hierarchical
Competition‐Colonization Models. The American Naturalist153 : 371–383. DOI: 10.1086/303182
Krull ES, SKJEMSTAD JO, BALDOCK JA. 2004. Functions of soil
organic matter and the effect on soil properties .
Lampurlanés J, Cantero‐Martínez C. 2003. Soil Bulk Density and
Penetration Resistance under Different Tillage and Crop Management
Systems and Their Relationship with Barley Root Growth. Agronomy
Journal 95 : 526–536. DOI:
https://doi.org/10.2134/agronj2003.5260
Le Bissonnais Y, Prieto I, Roumet C, Nespoulous J, Metayer J, Huon S,
Villatoro M, Stokes A. 2018. Soil aggregate stability in Mediterranean
and tropical agro-ecosystems: effect of plant roots and soil
characteristics. Plant and Soil 424 : 303–317. DOI:
10.1007/s11104-017-3423-6
Liu Z, Zhou W, Shen J, Li S, Ai C. 2014. Soil quality assessment of
yellow clayey paddy soils with different productivity. Biology and
Fertility of Soils 50 : 537–548. DOI:
10.1007/s00374-013-0864-9
Maurya S, Abraham JS, Somasundaram S, Toteja R, Gupta R, Makhija S.
2020. Indicators for assessment of soil quality: a mini-review.Environmental Monitoring and Assessment 192 : 604. DOI:
10.1007/s10661-020-08556-z
Moinet GYK, Midwood AJ, Hunt JE, Rumpel C, Millard P, Chabbi A. 2019.
Grassland Management Influences the Response of Soil Respiration to
Drought. Agronomy-Basel 9 : 124. DOI:
10.3390/agronomy9030124
Mujdeci M, Simsek S, Uygur V. 2017. The Effects of Organic Amendments on
Soil Water Retention Characteristics Under Conventional Tillage System.Fresenius Environmental Bulletin 26 : 4075–4081
Müller G, Gastner M. 1971. The’Karbonat-Bombe’, a simple device for the
determination of carbonate content in sediment, soils, and other
materials. Neues Jahrbuch für Mineralogie-Monatshefte10 : 466–469
Mulvaney RL. 1996. Nitrogen—Inorganic Forms. Methods of Soil
Analysis . John Wiley & Sons, Ltd, 1123–1184. DOI:
10.2136/sssabookser5.3.c38
Murphy DJ, Horner RM, Clark CE. 2015. The impact of off-site land use
energy intensity on the overall life cycle land use energy intensity for
utility-scale solar electricity generation technologies. Journal
of Renewable and Sustainable Energy 7 : 033116. DOI:
10.1063/1.4921650
Novara A, Armstrong A, Gristina L, Semple KT, Quinton JN. 2012. Effects
of soil compaction, rain exposure and their interaction on soil carbon
dioxide emission. Earth Surface Processes and Landforms37 : 994–999. DOI: 10.1002/esp.3224
Olmstead LB, Alexander LT, Middleton HE. 1930. A pipette method of
mechanical analysis of soils based on improved dispersion procedure .
Ong S, Campbell C, Denholm P, Margolis R, Heath G. 2013. Land-Use
Requirements for Solar Power Plants in the United States . National
Renewable Energy Lab. (NREL), Golden, CO (United States). DOI:
10.2172/1086349
Quinton JN, Govers G, Van Oost K, Bardgett RD. 2010. The impact of
agricultural soil erosion on biogeochemical cycling. Nature
Geoscience 3 : 311–314. DOI: 10.1038/ngeo838
Rabaia MKH, Abdelkareem MA, Sayed ET, Elsaid K, Chae K-J, Wilberforce T,
Olabi AG. 2021. Environmental impacts of solar energy systems: A review.Science of The Total Environment 754 : 141989. DOI:
10.1016/j.scitotenv.2020.141989
Raich JW, Tufekciogul A. 2000. Vegetation and soil respiration:
Correlations and controls. Biogeochemistry 48 : 71–90.
DOI: 10.1023/A:1006112000616
Raiesi F, Salek‐Gilani S. 2020. Development of a soil quality index for
characterizing effects of land-use changes on degradation and ecological
restoration of rangeland soils in a semi-arid ecosystem. Land
Degradation & Development 31 : 1533–1544. DOI:
https://doi.org/10.1002/ldr.3553
Ranjard L, Dequiedt S, Jolivet C, Saby NPA, Thioulouse J, Harmand J,
Loisel P, Rapaport A, Fall S, Simonet P, Joffre R, Bouré NC-P, Maron
P-A, Mougel C, Martin MP, Toutain B, Arrouays D, Lemanceau P. 2011.
Biogeography of Soil Microbial Communities: A Review and a Description
of the Ongoing French National Initiative. Sustainable Agriculture
Volume 2 857–865. DOI: 10.1007/978-94-007-0394-0_37
Romero-Díaz A, Ruiz-Sinoga JD, Robledano-Aymerich F, Brevik EC, Cerdà A.
2017. Ecosystem responses to land abandonment in Western Mediterranean
Mountains. CATENA 149 : 824–835. DOI:
10.1016/j.catena.2016.08.013
Rutgers M, Schouten AJ, Bloem J, Eekeren NV, Goede RGMD, Akkerhuis
GAJMJ, Wal AV der, Mulder C, Brussaard L, Breure AM. 2009. Biological
measurements in a nationwide soil monitoring network. European
Journal of Soil Science 60 : 820–832. DOI:
https://doi.org/10.1111/j.1365-2389.2009.01163.x
Saetre P. 1998. Decomposition, Microbial Community Structure, and
Earthworm Effects Along a Birch–Spruce Soil Gradient. Ecology79 : 834–846. DOI:
10.1890/0012-9658(1998)079[0834:DMCSAE]2.0.CO;2
Saiya-Cork KR, Sinsabaugh RL, Zak DR. 2002. The effects of long term
nitrogen deposition on extracellular enzyme activity in an Acer
saccharum forest soil. Soil Biology and Biochemistry 34 :
1309–1315. DOI: 10.1016/S0038-0717(02)00074-3
Scarpare FV, van Lier Q de J, de Camargo L, Pires RCM, Ruiz-Correa ST,
Bezerra AHF, Gava GJC, Dias CTS. 2019. Tillage effects on soil physical
condition and root growth associated with sugarcane water availability.Soil & Tillage Research 187 : 110–118. DOI:
10.1016/j.still.2018.12.005
Schimel J, Schaeffer SM. 2012. Microbial control over carbon cycling in
soil. Frontiers in Microbiology 3 . DOI:
10.3389/fmicb.2012.00348
Schnürer J, Rosswall T. 1982. Fluorescein Diacetate Hydrolysis as a
Measure of Total Microbial Activity in Soil and Litter. Applied
and Environmental Microbiology 43 : 1256–1261
Shao P, He H, Zhang X, Xie H, Bao X, Liang C. 2018. Responses of
microbial residues to simulated climate change in a semiarid grassland.Science of the Total Environment 644 : 1286–1291. DOI:
10.1016/j.scitotenv.2018.07.055
Simansky V, Bajcan D, Ducsay L. 2013. The effect of organic matter on
aggregation under different soil management practices in a vineyard in
an extremely humid year. Catena 101 : 108–113. DOI:
10.1016/j.catena.2012.10.011
Six J, Bossuyt H, Degryze S, Denef K. 2004. A history of research on the
link between (micro)aggregates, soil biota, and soil organic matter
dynamics. Soil and Tillage Research 79 : 7–31. DOI:
10.1016/j.still.2004.03.008
Sparling G. 1997. Soil microbial biomass, activity and nutrient cycling
as indicators of soil health. Biological indicators of soil
health 97–119
Tabatabai MA, Bremner JM. 1969. Use of p-nitrophenyl phosphate for assay
of soil phosphatase activity. Soil biology and biochemistry1 : 301–307
Tanner KE, Moore‐O’Leary KA, Parker IM, Pavlik BM, Hernandez RR. 2020.
Simulated solar panels create altered microhabitats in desert landforms.Ecosphere 11 : e03089. DOI: 10.1002/ecs2.3089
Team RC. 2020. R: A language and environment for statistical computing.
Telak LJ, Bogunovic I. 2020. Tillage-induced impacts on the soil
properties, soil water erosion, and loss of nutrients in the vineyard
(Central Croatia). Journal of Central European Agriculture21 : 589–601. DOI: 10.5513/JCEA01/21.3.2810
Velasquez E, Lavelle P, Andrade M. 2007. GISQ, a multifunctional
indicator of soil quality. Soil Biology and Biochemistry39 : 3066–3080. DOI: 10.1016/j.soilbio.2007.06.013
Wall DH, Bardgett RD, Behan-Pelletier V, Herrick JE, Jones TH, Six J,
Strong DR, Putten WH van der, Ritz K. 2012. Soil Ecology and
Ecosystem Services . OUP Oxford
Yang L, Liu N, Ren H, Wang J. 2009. Facilitation by two exotic Acacia:
Acacia auriculiformis and Acacia mangium as nurse plants in South China.Forest Ecology and Management 257 : 1786–1793. DOI:
10.1016/j.foreco.2009.01.033
Yin R, Kardol P, Thakur MP, Gruss I, Wu G-L, Eisenhauer N, Schaedler M.
2020. Soil functional biodiversity and biological quality under threat:
Intensive land use outweighs climate change. Soil Biology &
Biochemistry 147 : 107847. DOI: 10.1016/j.soilbio.2020.107847
Zhang Y, Xu X, Li Z, Liu M, Xu C, Zhang R, Luo W. 2019. Effects of
vegetation restoration on soil quality in degraded karst landscapes of
southwest China. Science of The Total Environment 650 :
2657–2665. DOI: 10.1016/j.scitotenv.2018.09.372
Zou CB, Barnes PW, Archer S, McMurtry CR. 2005. Soil moisture
redistribution as a mechanism of facilitation in savanna tree–shrub
clusters. Oecologia 145 : 32–40. DOI:
10.1007/s00442-005-0110-8
Table
Table 1 Environmental and technical characteristics of solar parks.