Full title:

EFFECTS OF TOPOGRAPHY AND SURFACE SOIL COVER ON EROSION FOR

MINING RECLAMATION. THE EXPERIMENTAL SPOIL HEAP AT EL MACHORRO
MINE (CENTRAL SPAIN)

Short title:
TOPOGRAPHY AND SURFACE SOIL COVER IN MINING RECLAMATION

MARTÍN-MORENO, C.1*, MARTÍN DUQUE, J.F.1, NICOLAU IBARRA, J.M.2, HERNANDO

RODRÍGUEZ, N.3, SANZ SANTOS, M.A.3, AND SÁNCHEZ CASTILLO, L.4

1
Department of Geodynamics, Complutense University of Madrid and Institute of Geosciences - IGEO (CSIC,UCM),
C/José Antonio Novais 2, Madrid E-28040, Spain.
2
Department of Agricultural and Environmental Sciences, University of Zaragoza, E-22071 Huesca, Spain.
3
Department of Geodynamics, Complutense University of Madrid, C/José Antonio Novais 2, Madrid E-28040, Spain.
4
Department of Geologic Engineering, Polytechnic University of Madrid, C/Ríos Rosas 21, Madrid E-28003, Spain.

* Correspondence to: C. Martín Moreno, Department of Geodynamics, Complutense University of Madrid and Institute of
Geosciences - IGEO (CSIC,UCM), C/José Antonio Novais 2, Madrid E-28040, Spain.
E-mail: [email protected]; phone, 34-913944676; fax, 34-91-3944845.

This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article
as doi: 10.1002/ldr.2232

This article is protected by copyright. All rights reserved.

 ABSTRACT

Mining reclamation tries to reduce environmental impacts, including accelerated runoff, erosion and
sediment load in the nearby fluvial networks and their ecosystems. This study compares the effects of
topography and surface soil cover on erosion on man-made slopes coming from surface mining
reclamation in Central Spain. Two topographic profiles, linear and concave, with two surface soil
covers, subsoil and topsoil, were monitored for two hydrologic years. Sediment load, rill
development, and plant colonization from the four profiles were measured under field conditions. The
results show that, in the case of this experiment, a thick and non-compacted topsoil cover on a linear
slope yielded less sediment than carbonate colluvium or topsoil cover on a concave slope. This study
also shows that vegetation establishment, which plays an important role in erosion control, depends
on topography. Plant cover was more widespread and more homogeneous on linear profiles with
topsoil cover. On concave slopes, plant establishment was severely limited on the steepest upper part
and favoured in the bottom. This study suggests that management of topography and surface soil
cover should be approached systematically, taking three outcomes into consideration: i) topsoil can
lead to a successful mining reclamation regardless of topography, ii) created concave slopes can lead
to a successful mining reclamation, and iii) topography determines the vegetation colonization
pattern.

Key words: topographical design, topsoil, constructed slopes, concave slopes, water erosion,
vegetation.

INTRODUCTION

Mining, which supplies materials thought essential for our society, has serious environmental impacts.
Opencast mining impacts all ecosystem components: substrata, topography, hydrology (surface and
groundwater), soil, vegetation, fauna, atmosphere, and landscapes (Osterkamp & Morton, 1996;
Evans, 2000; Rivas et al., 2006). Often, mining impacts also have adverse effects on nearby
ecosystems. Among these off-site effects, the hydrologic impact of mines on downstream fluvial
ecosystems is one of the most detrimental (Toy & Hadley, 1987; Nicolau & Asensio, 2000).

Theoretically, mining reclamation should reduce these impacts. However, in spite of the significant
development of mining reclamation techniques over the years, failures on mining reclamation are
common (Haigh, 2000). Inadequate management of landform design at many reclaimed mining sites
has been identified as the main reason for reclamation failures because of accelerated water erosion
(Loch, 1997; Nicolau & Asensio, 2000).

This article is protected by copyright. All rights reserved.

To achieve effective control of water erosion, an integrated management of topography, surface soil
cover, and vegetation is required (Nicolau, 2003). Of these three factors, the management of
topography and surface soil cover is considered an essential component of mining reclamation
practices by many (e.g., Evans & Willgoose, 2000; Toy & Black, 2001; Moliere et al., 2002; Toy &
Chuse, 2005).

For mine reclamation to be successful, efforts also must be directed towards the creation of
biologically functional and stable soils that reduce soil erosion and facilitate the rehabilitation of post-
mined lands (Bradshaw & Chadwick, 1980; Whisenant et al., 1995). Soil erosion negatively affects
vegetation growth through several mechanisms: the removal of seeds and nutrients from surface soil,
direct plant removal, and the loss of water through surface runoff (Pimentel et al., 1995; Espigares et
al., 2011). Indeed, seeds removal is sometimes a negligible reason to explain the lack of vegetation
even in bare surfaces (see Cerdá & García-Fayos, 1997). The most common soil surface used is
topsoil (coversoil) spread on the slope surface; this approach is considered essential in most cases
(Power et al., 1981; Kapolka & Dollhopf, 2001). Additionally, a wide range of modifications can be
applied to improve physical and chemical soil properties (Bradshaw & Chadwick, 1980). Armoring
surface with rocks is a convenient and cost-effective measure to decrease soil erodibility (Toy et al.,
2002).

The most common approach of topography management consists of terraced landforms, graded spoil
banks comprising alternating short constant-gradient slopes and benches. Artificial ditches commonly
drain off the concentrated runoff (Bugosh, 2006). Without maintenance, many terraced landforms
succumb to water erosion in the long term (Loch, 1997). Linear slopes can be unstable, especially if
the base level is continuously changing by ditch incision, which causes the slopes to respond by
eroding or mass failure (e.g. Haigh, 1980, 1985). Erosion problems also arise due to ponding or
exceeding the storage capacity of the terraces (Sawatsky et al., 2000). According to Hancock et al.
(2003), linear slopes erode and increase sediment loss until achieving a stable profile, which is usually
concave. Additionally, we have reported how terraced spoil heaps in this physiographic setting of the
Upper Tagus are not stable within a decadal span time, and they evolve to gullied landforms (see Sanz
et al., 2008).

Arguments have frequently been raised in favour of topographic designs that replicate ‘natural’
landscapes. This geomorphic approach is based on knowledge of geomorphic processes, mostly
fluvial processes operating for an extended period of time. The objective of these designs is the
construction of steady-state landscapes (Riley, 1995; Schor & Gray, 2007).

This article is protected by copyright. All rights reserved.

Application of truly geomorphic approaches (Sawatsky & Beckstead, 1996; Toy & Chuse, 2005)
depends very much on the exploitation method and timing. Implementing a geomorphic approach is
more difficult and expensive in active mines which already have terraced landforms. Often, only basic
modifications of individual slopes (contour berm or contour linear steep slope) can be cost-effective.
Geomorphic approaches are easier to implement before mining activities start or at abandoned mines.
These two situations highlight the success of Bugosh’s approach, a computerized method (GeoFluv)
of mining reclamation based on fluvial geomorphic principles (Bugosh, 2004). His approach seeks
hydrologic balance in reclaimed minescapes and is perfectly tuned with the approach of Toy & Chuse
(2005) who suggested that constructed landscapes should include hydrologic basins, composed of
slopes and watercourses. When basic modification of individual slopes is the only possibility, the
GeoFluv method plays an important role to decrease the slope length factor. This is carried out by
building first and second order channel drainage density, so that frequent small subwatersheds
transform long slopes in shorter ones, making the resultant landforms more resistant to erosion.

The topographic profile of individual constructed slopes has been discussed for long in the field of
mining reclamation (Haigh, 1985; Toy et al., 2002; Hancock et al., 2003). Many studies have reported
a relationship between soil erosion and slope shape. These include the first studies in geomorphology
related to soil erosion on individual slopes (Meyer & Kramer, 1969), laboratory experiments (D'Souza
& Morgan, 1976), and the application of erosion models. For example, Hancock et al. (2003) and
Priyashanta et al. (2009) applied the SIBERIA model to demonstrate the greater stability of concave
slopes compared to linear ones. However, no field experimental studies have been conducted to assess
the reclamation benefits of concave slopes compared to linear slopes.

Because less sediment exportation occurs on concave slopes compared to other shapes (linear, convex
or S-shape) (Meyer & Kramer, 1969), these studies have led to the belief that concave slopes are very
stable. While watershed size and runoff increase downslope, the slope gradient decreases, and this
reduces runoff velocity and erosion ability (Toy et al., 2002).

Martín-Duque et al. (2010) explained how a holistic geomorphic approach to mining reclamation,
using both topographic and surface soil cover management, led to a successful mining reclamation in
a quarry of Central Spain. The current study is based on that work and describes a field experiment
carried out at the El Machorro kaolin mine of Central Spain. The objective of this study was to
compare the erosion response of two constructed slopes, linear and concave, with two different
surface soil covers. These soil covers were: i) subsoil (carbonate colluvium), a natural superficial
sediment that drapes the sandy sedimentary rocks underlying the original slopes around the mine, and
ii) topsoil, soils developed originally on top of the carbonate colluvium. A linear slope of overburden
material with no cover was used as a control for linear slopes. A concave slope of overburden material

This article is protected by copyright. All rights reserved.

with no cover could not be constructed, because the experimental layout had to be adapted to pre-
existing topographic conditions. Therefore, a total of four different combinations of topography and
surface soil cover, that we call ‘reclamation treatments’, and one control (overburden linear slope),
were monitored in this study. A core objective of this study was to compare the response of both
topographies and both surface soil covers, to acquire know-how for efficient mining reclamation at
similar sites. Our working hypothesis was that concave slopes would yield less sediment than linear
slopes. We also expected a dramatic reduction in soil loss from topsoil and carbonate colluvium
compared to overburden material.

MATERIALS AND METHODS

Study area

El Machorro is an active contour mine with an ongoing terraced reclamation approach. It is located in
the buffer zone of the Upper Tagus Natural Park (UTNP, Parque Natural del Alto Tajo, in Spanish) in
Central Spain (40º 39’ 29” N, 2º 2’ 26” W, datum World Geodetic System 1984, WGS84) (Figure 1).
This protected area was established in 2000 by a regional law (DOCM, 2000) because of its
outstanding biodiversity, specifically regarding aquatic ecosystems. It is also very diverse
geologically (Carcavilla et al., 2008) and biologically (DOCM, 2000).

The Upper Tagus landscape is characterized by plateaus and mesas capped by Cretaceous carbonates,
with their slopes and canyon scarps underlain by sandy sediment that hold the kaolin (Arenas de
Utrillas Formation) exploited in several mines (Olmo & Álvaro, 1989; González Amuchastegui,
1993).

On mesa tops, the soils are chromic luvisols, calcaric cambisols, mollic leptosols, and rendzic
leptosol. On slopes, carbonate colluvia with calcaric cambisols are common (IUSS Working Group
WRB, 2007). The vegetation is representative of mediterranean-continental environments, with
communities dominated by Juniperus thurifera on the highest plateaus, and pine (Pinus nigra subsp.
salzmanii) and gall oak (Quercus faginea) in valleys (MARM, 1997–2006).

The climate of this area is temperate mediterranean with dry and mild summers (Csb, according to
Köppen, 1918), but with a noticeable continental influence. The moisture regime is dry mediterranean
(Papadakis classification) (CNIG, 2004). Mean annual precipitation is 780 mm and mean annual
temperature is 10ºC (AEMET, 2012). Seasonally, this area is characterized by long and cold winters
with snow common and short, dry summers with high intensity rainstorms. The spring and fall are

This article is protected by copyright. All rights reserved.

usually wet. The rainfall erosive factor, R (equivalent to the R factor of RUSLE), is estimated to be
about 80 (ICONA, 1988).

Rainfall and temperature monitoring

To measure rainfall quantity and intensity, a tipping-bucket automatic raingauge (0.2 mm/pulse)
(Davis Instruments, 2005) with a HOBO Event data logger was installed 100 m away from the
experimental spoil heap, at 1230 m.a.s.l. Raingauge data were downloaded at the same time as the
sediment collection. Total rainfall volume (mm) and maximum rainfall volume in 24 hours (mm)
were calculated. In addition, the return period of annual precipitation for each year was estimated
using the CHAC software (CEDEX, 2004). Each year, temperature data were obtained from a nearby
weather station (AEMET, 2012).

Experimental design

An experimental spoil heap was built by the mining operator company of El Machorro mine,
CAOBAR, in the summer of 2008, on the foundations of an existing spoil heap. Two different
topographic slope shapes, linear and concave, were constructed with spoils (overburden materials)
and covered with two surface soil covers: subsoil (carbonate colluvium), and topsoil (Table I).
Additionally to these four reclamation treatments, one linear slope of the spoil heap with overburden
material (spoils) was left uncovered as a control (Figure 2). The four ‘reclamation treatments’ and the
control were monitored for two hydrologic years (2009 and 2010) starting from November 6, 2008.

At the experimental spoil heap, articulated dump trucks built the terraced spoil heaps by directly
unloading materials, and a bulldozer compacted and finished the benches. The dump trucks could not
drive on the linear slopes because of their high slope gradient, so the trucks drove on the benches and
unloaded the two surface soil covers directly downslope. The concave slope was built by a bulldozer
that drove on the concave slope reshaping it and spreading the surface soil covers at the same time.
Summing up, the experimental spoil heap had two parts. The first one was a terraced system with two
linear slopes and one intermediate bench. Each linear slope had the two surface soil covers (carbonate
colluvium and topsoil) and the exposed overburden material (control); the second part was a concave
slope with the two surface soil covers, therefore five different slopes were monitored (see Figure 2).

Mining and reclamation operations within the mine prevented the construction of the upper part of the
concave slope during the first hydrologic year of the study. During this period, the concave slope
consisted of its half-lower part, equivalent in height to a single linear slope plus its bench.
Additionally, run-off from the upper slope formed an alluvial fan on the concave slope covered with

This article is protected by copyright. All rights reserved.

carbonate colluvium. Therefore, data could not be collected on this treatment during the first year.
The concave slope was fully constructed in the second year to have the same width and length as a set
of two linear slopes with an intermediate bench. This modification could be considered a limitation of
this study.

Linear slopes had a mean length of 11 m (standard deviation 0.6), with a slope gradient of 32º. The
bench was 5 m wide with a reversed-slope gradient of 14º in cross section and 2º in longitudinal
section. Concave slopes had a slope length of 25 to 30 m during the first year and 35 to 40 m during
the second year. Their gradient increased from bottom to top from 4º to 26º (first year) and from 4º to
32º (second year) (See table II for details). The concave slope curvature was described using the
equation proposed by Stefano et al. (2000):
n
 x
y = H 1 − 
 λ

where x = horizontal abscissa and y = the corresponding elevation

H= difference of level
λ = slope length measured along the horizontal axis
n= exponent that varies according slope shape, following Stefano et al. (2000)

Short concave slopes (first year) had an n value between 1.34 and 1.32, whereas long concave slope
values (second year) were between 1.40 and 1.47 (Figure 3). A differential Global Positioning System
(GPS, model number Leica 1200) was used to survey the concave slope profiles. Slope surveys were
conducted once a year (12 May 2009 and 1 July 2010).

Three composite samples were taken from each soil cover to characterize their physical properties
(shown in Table III). The thickness of both carbonate colluvium and topsoil ranged between 30 and
75 cm on linear slopes. This wide range resulted from directly unloading material from upslope
without spreading it. Carbonate colluvium and topsoil on concave slopes were 20-30 cm thick, and
were spread by a bulldozer.

The core of this study is based on the field measurement of the sediment amount yielded by each
reclamation treatment and the control. Three open plots were set up for every slope. Sediment amount
was recorded using silt fences (Robichaud & Brown, 2002), with a width of 3 m, placed across the toe
of the slopes. Silt fences trap sediment while allowing water to pass through. According to Robichaud
& Brown (2002), the trap efficiency of silt fences is 68 to 98%. Because sediment could fill and

This article is protected by copyright. All rights reserved.

overload silt fences, possibly resulting in a loss of sediment, periodic cleaning of silt fences was
necessary (Robichaud & Brown, 2002).

Sediment yield was measured at the toe of the concave slope and at the toe of the lower single linear
slope of the set of two linear slopes (Figure 2). Sediment from the upper linear slope were not
measured, but they did not run onto the monitored lower linear slope, as they were deposited on the
intermediate reversed sloped bench and drained out of the monitored lower linear slope (Figure 2).
The short reversed slope of the terrace bench was not counted in the balance, as it was observed that it
did not yield any sediment.

Therefore, a total of 12 (first year) and 15 (second year) sets of ‘open’ plots (plots without artificial
boundaries) with silt fences were monitored. Since the plots were open, there were differences in plot
size due to different drainage areas. The area of each open plot, measured using differential GPS,
ranged between 23.5 and 83.7 m2 (first year) and between 23.5 and 124 m2 (second year) (Table II).

Sediment yield

The protocol for monitoring the open plots consisted of collecting the sediment trapped by the silt
fences and weighing the sediment in the field, using a portable weight scale. The sediment from a
single plot was mixed and a portion of the mixed sediment was taken to calculate the percentage of
moisture, using the method by Ramos-Scharrón & McDonald (2007). The erosion rate was calculated
and the results were expressed as Mg ha-1yr-1. Annual sediment yields and standard deviations were
also calculated for each treatment.

Rill development

Overburden materials at El Machorro mine are mainly sandy, with very low clay content. The very
low cohesion makes the overburden material vulnerable to detachment by runoff, so that gully
formation is common.

To monitor the landform evolution of the four different reclamation treatments and the control,
photographs were taken of each open plot before sediment was collected. Rill networks were
measured after they formed. Width and depth were measured in at least 80% of all rills in three slope
positions (top, middle, and bottom).

The length, width, and depth of rills and gullies were measured with a tape, following the method
described by Morgan (Morgan, 2005). Rill volume was estimated by multiplying the rill cross-

This article is protected by copyright. All rights reserved.

sectional area —“U” shape, for carbonate colluvium and “V” shape, for overburden material— by
their mean rill length. This rill volume was then divided by the treatment area, to obtain estimated
values for sediment removed by rill erosion (m3 m-2). This value was then transformed to sediment
weight per area (Mg ha-1) by multiplying the volume by the mean bulk density of each surface soil
cover that was calculated by the core method (Sobek et al., 1978). Three soil core samples were taken
from each slope treatment for bulk density calculations. The sediment amounts resulting from rill
measurements and from the silt fences were then compared.

Vegetation colonization

Vegetation cover was measured using digital photographs and a point-frequency method
(Brakenhielm & Liu, 1995; Vanha-Majamaa et al., 2000) one year after the end of the second year of
the study (October 2011). Because no seeding was applied in any of the reclamation treatments, we
therefore measured spontaneous vegetation colonization.

Statistical analysis

To compare the effects of topography and surface soil cover on sediment yield, paired t-tests were
conducted comparing sediment yield from treatments with the same topography but with different
surface soil cover (i.e. linear slope with carbonate colluvium vs linear slope with topsoil) and
sediment yield from treatments with same surface soil cover but with different topography (i.e.
concave slope with topsoil vs linear slope with topsoil). Analyses were conducted separately for each
study year. For linear slopes, data were also analyzed for both years combined, because the plots were
not modified during the second year. Statistical analyses were made using Statgraphics Centurion
XVI.I software, version 16.1.17 (StatPoint Technologies Inc., 2012). The significance level was
α=0.05.

RESULTS

Rainfall and temperature

A total of 324 rain days were registered during the study period, accounting for a total rainfall of 1426
mm. Annual rainfall for the second year (992 mm) was approximately twice that of the first year (434
mm), with return periods of 5 and <2 years, respectively. Climatic characteristics of each study year
are shown in Table IV. Monthly rainfall ranged from 1 mm (July 2009) to 290 mm (December 2009).
The maximum rainfall recorded in 24 hours was 49 mm. Frost-free days were slightly more common
in the second year.

This article is protected by copyright. All rights reserved.

Sediment yield

During the two years studied, open plots were sampled approximately once a month, resulting in a
total of 21 samples: 10 samples during the first year and 11 samples during the second year. Mean
sediment yield and standard deviation of each reclamation treatment are shown in Table V, along with
rainfall characteristics for the period between two consecutive sediment collections. The sediment
yield rates for the three plots within the same treatment did not differ significantly (p>0.05, paired t-
test).

Significant differences were found when sediment yield rates from reclamation treatments with the
same topography but different surface cover were compared (Table VI). For the first year, the
comparison between linear slope with topsoil (LS-TS) and linear slope with overburden material (LS-
OM) showed a significant difference (p=0.01, t-test). For the second year, the comparison of these
two treatments also showed a significant difference (p=0.003). Regarding the two-year data analyses,
significant differences were found between all tested pairwise treatments on linear slopes (p<0.05,
paired t-test). When slopes with the same surface cover but different topography were compared, no
meaningful significant differences were found.

Regarding annual sediment yield rates, the short concave slope with topsoil (SCS-TS) had lower
sediment yield values than any linear slope during the first year, regardless of surface soil cover
(Figure 4 and Table V). The sediment yield rates of linear slopes depended on the surface soil cover:
the slope with topsoil had the lowest rate (12 Mg ha-1yr-1), one order of magnitude less than that with
carbonate colluvium (120 Mg ha-1yr-1) or overburden material (282 Mg ha-1yr-1). In the second year,
the linear slope with topsoil (LS-TS) produced the lowest erosion rate (3 Mg ha-1yr-1). The other two
linear slopes had the higher values: 126 Mg ha–1yr–1 with carbonate colluvium and 347 Mg ha–1yr–1
with just overburden. The effect of surface soil cover was not found for the long concave slopes. The
slope with topsoil (LCS-TS) yielded 20 Mg ha-1yr-1 of sediment and the slope with carbonate
colluvium (LCS-CC) yielded 16 Mg ha-1yr-1 (Figure 4).

Rill development

Rill development was different on concave and linear slopes. Concave slopes developed a rill network
in the upper part, lacking rills in its lower part. Linear topography allowed a continuous rill network
along the slope length. In both cases, rill development depends on the surface soil cover
characteristics.

This article is protected by copyright. All rights reserved.

Rill development on concave slopes

The concave slope covered with topsoil (SCS-TS) did not develop rills during the first year, which
was dryer than the second one. Indeed, this treatment resisted the most intense rainfall in 24 hours of
the first year (38.4 mm), which occurred just after building the experimental spoil heap and spreading
the surface soil cover, but before the silt fences were installed. During the second year, small rills
formed in the steepest area of the concavity, near the top of the slope, but they were small and
disappeared downslope. These rills were not measured, because we assumed the sediment eroded
from them was deposited within the slope.

Plots on the concave slope with carbonate colluvium surface soil cover (SCS-CC) could not be
monitored during the first year, because run-on from upslope formed noticeable alluvial cones within
the open plots. In the second year, the upper parts of both concave slopes were reconstructed, making
them longer. During the second year, the concave slopes behaved similarly, regardless of their surface
soil cover: rills were formed at the top of the slope and disappeared downslope. On the long concave
slope with carbonate colluvium, these rills were discontinuous, with a “U” shape, and mean length of
6 m. The estimated sediment volume eroded from these rills over the two-year period was 1.4 m3, or
0.004 m3 m-2, based an area of 330 m2 on the LCS-CC. No mass movements, such as mudflows,
occurred on the concave slope with carbonate colluvium. The calculated bulk density for carbonate
colluvium was 1.26 g cm-3, so the estimated weight of sediment from the concave slope with
carbonate colluvium was 50 Mg ha-1. Since 80% of rills were measured, the estimated total mass of
sediment was 63 Mg ha-1. Two-year sediment yield measured in the open plots of this same slope was
16 Mg ha-1. The estimated amount of sediment determined from rill development has the same order
of magnitude as that measured at the silt fences, for the two-year period (Figure 6).

Rill development on linear slopes

The linear slope with topsoil (LS-TS) did not develop perceptible erosive forms during the two years.
The linear slope covered with carbonate colluvium (LS-CC) was subject to small mudflows in the first
year. Additionally, an incipient rill network developed. After this initial geomorphic evolution, the
plots remained very stable throughout the two-year period, with only small mudflows and minor rills.
At the end of the second year, rills were discontinuous, with a “U” shape, with an average width of 30
to 40 and depth of 10. The estimated average length was 7 m, and the estimated sediment volume
eroded from rills was 0.4 m3. The estimated sediment removed by rill erosion was 0.004 m3 m-2.
Considering the corresponding bulk density (1.27 g cm-3), the estimated sediment yield was 51 Mg ha-
1
(from 80% of rills), corresponding to a total sediment of 64 Mg ha-1 (for 100%). This estimated

This article is protected by copyright. All rights reserved.

sediment yield is one order of magnitude lower than that measured at the silt fences (246 Mg ha-1 for
the two-year period) (Figure 6).

The linear slope covered with overburden material (LS-OM) developed an evenly defined rill
network. These rills were deeper and much more numerous than those formed on the carbonate
colluvium. The rills were 20 cm-wide on average, and had an average depth of 20 to 30 cm, maximum
50 cm, at the end of the first year (Figure 5). Small alluvial cones were formed at the bottom of the
slopes. A progressive disintegration of sand clods on the linear slope surface was also observed during
the two years. During the second year, the rill-erosion process continued, leading to the formation of
gullies, being these landforms defined in the same way that Brice (1966, p. 290): “a recently extended
drainage channel that transmits ephemeral flow, has steep sides, a steeply sloping or vertical head
scarp, a width greater than about 1 foot, and a depth greater than about 2 feet”. At the end of the
second year, the rills were continuous, “V”-shaped, with an average width and depth of 45 cm and 25
cm, respectively. Gullies with a maximum width of 200 cm and depth of 150 cm were also measured.
Rill length was the same as on the linear slope, 11 m. The estimated sediment volume eroded from
rills was 4.75 m3, and 0.045 m3 m-2, the highest of the slopes monitored (Figure 6). The estimated
sediment eroded by rill processes, calculated using the bulk density of 1.41 g cm-3, was 793 Mg ha-1
(considering 100% of rills). The estimated sediment yield quantified from rill development was higher
than that measured at the silt fences (629 Mg ha-1 for the two-year period).

Vegetation colonization

At the start of the study period all plots were bare, without any vegetation. As geomorphic evolution
progressed, natural plant colonization occurred. Concave and linear slopes covered with topsoil
showed plant establishment in the following spring (spring of 2009). In October 2011, plants covered
30% of the concave slope and 50% of the linear slope (table III). Plants spatial pattern was not
homogeneous on the concave slope with topsoil, so that plants were not evenly distributed along the
slope, but the linear slope showed a uniform vegetation distribution. On the concave slope, vegetation
cover was more extensive in the lower part of the slope than at the top. Table VII shows the plant
species identified in each topsoil-covered slope. Although species richness is similar in both slopes
(14), species composition is quiet different (being only 5 species common among to the two slopes).
No vegetation was observed on carbonate colluvium or overburden material.

This article is protected by copyright. All rights reserved.

 DISCUSSION
Sediment yield

Our results suggest that surface soil cover controls sediment yield on linear slopes more than on
concave ones. This is supported by the fact that linear topography has no mechanisms to control
sediment fluxes, while concave topography is able to store sediment at the toe (Stefano et al., 2000;
Toy et al., 2002). On linear slopes, control of erosion could be improved by using a different surface
soil cover. Our results are consistent with previous findings: topsoil was the best surface soil cover,
providing better conditions for soil development and plant establishment than other materials (Power
et al., 1981; Haigh, 2000).

Similar erosive response was observed in the first year for the topsoiled slopes, whether short concave
(SCS-TS) or linear (LS-TS), indicating that, under favorable soil conditions, the role of topography
was not evident. During the second year, topsoiled slopes behaved differently. While sediment yield
from the linear slope with topsoil (LS-TS) was reduced, sediment yield from the long concave slope
(LCS-TS) was greater than the yield from the short concave slope (SCS-TS). The increased length
and drainage area could explain the increase in sediment yield. In agreement with this, several authors
have reported that, under the same environmental conditions, shorter slopes produce less sediment
than longer ones (Toy & Foster, 1998; Liu et al., 2000; Toy et al., 2002; Toy & Chuse, 2005).

Another aspect must be considered: constraints existed for combining soil surface covers and
topography. The depth, uniformity, and quality of surface soil cover were determined by reclamation
operations. On linear slopes, the surface soil cover was spread out by direct unloading of trucks,
which provided a more homogeneous and less compacted layer. However, on concave slopes the
spreading out process had to be carried out with a bulldozer, which compacted the soil (Barber &
Romero, 1994; Chong & Cowsert, 1997). Soil compaction has been reported to reduce the land’s
capacity to absorb rainwater, accelerating runoff and erosion (Haigh & Sansom, 1999). The greater
thickness and porosity of linear slopes with topsoil, as well as a better spatial distribution of surface
soil cover, could explain lower rates of sediment yield than for the concave slope. This means that
slope topography affects surface soil cover depth and quality in reclaimed landscapes (Hancock et al.,
2003; Priyashanta et al., 2009) (see table VIII).

The smaller second-year sediment yield from the long concave slope (16 Mg ha-1yr-1) compared with
linear slope with carbonate colluvium (126 Mg ha-1yr-1) suggests that concave topography helps to
reduce sediment yield. The yield was smaller even though the concave slope was longer than the
corresponding linear slope, and even though the concave slopes had been recently constructed.

This article is protected by copyright. All rights reserved.

To assess the validity of the sediment yield measurements, it is important to take into account that,
although the plots were open, the length and area of the linear slopes were similar. Because of this, we
consider that converting sediment yield to per unit area, and comparing them, was justified. However,
the long concave slopes had larger open plots. A larger contributing area implies a higher erosive
power, but, the fact that the slope was concave implies a lower erosive power. The combined
consequence of these effects could not be separated and quantified. Therefore, converting sediment
yield to per unit area for concave slopes, and comparing them with linear slopes, has an evident
uncertainty. Despite of that, the comparison was made because they are real alternatives of
reclamation, both for this site and elsewhere: concave slopes or terraced ones as a topographic
possibility of regarding spoil heaps.

Rill development

In our experiment, rill development on linear slopes showed clear differences depending on the
surface soil cover. Whereas no rills were formed on the linear slope with topsoil (LS-TS), a
widespread rill network was developed on overburden material (LS-OM), and only few rills and
mudflows occurred on carbonate colluvium (LS-CC). This very different geomorphic behavior
indicates that soil cover is dominant in controlling erosion processes on linear slopes. Topsoil resists
erosion (Sawastky et al., 1996), because its higher infiltration rate decreases runoff and, therefore, soil
detachment (Haigh & Samson, 1999). On the other hand, rill erosion is very common in overburden
materials, because higher bulk density promotes overland flow (Soulliere & Toy, 1986; Moreno-de las
Heras et al., 2010). Two additional factors favored rill formation in overburden material: the low rock
cover and the sandy texture (Quansah, 1981; Porta et al., 1989) as described in table III.

Generally speaking, rills grow by incision and by side-wall sliding (Nicolau, 2002). The different
cross sections —V vs U shape— and size could be explained as a consequence of different surface
soil covers. Rills developed on overburden material were V-shaped and larger than those on carbonate
colluvium. This was likely due to the sandy texture and lower cohesion of overburden, favoring more
effective incision and side-wall collapse, and causing rill widening. Rills developed on carbonate
colluvium were observed to be U-shaped and smaller. This could be interpreted as a result of higher
cohesion in carbonate colluvium because of lower sand and higher silt content than in overburden
material. The carbonate colluvium also has a higher surface roughness (due to the abundance of rock
fragments), which would also contribute to a smaller rill size development. Roughness decreases
overland flow and runoff because of surface ponding and increased hydraulic roughness that reduces
the effective flow shear stress (Darboux et al., 2002; Toy et al., 2002; Gómez & Nearing, 2005).

This article is protected by copyright. All rights reserved.

Sediment yield estimated to have been eroded from rills differed from sediment yield measured in silt
fences. At least two factors affect the interpretation of the results. Sediment yield estimated from rills
assessment represented only rill erosion. For all comparisons it is important consider that rill
assessment has some limitations, and it is an estimation. At the same time, silt fences trap sediment
from rill, inter-rill erosion and mudflows, and it is necessary to take into account how efficiently the
silt fences trap sediment. According to Robichaud & Brown (2002) the total values for sediment yield
could be 2% to 32% higher. One might expect then that rill erosion estimates were probably low and
silt fence measurements could be higher.

For the linear slope with carbonate colluvium (LS-CC), sediment yield estimated from rills
assessment was one order of magnitude lower than sediment yield measured at silt fences (64 Mg ha-1
and 246 Mg ha-1 respectively, figure 6). This difference could be explained by the fact that small
mudflows occurred on this slope. For the linear slope with overburden material (LS-OM), the
estimated sediment yield from rills was 164 Mg ha-1 (21 %) higher than the sediment yield measured
in silt fences. This could be explained by the fact that small alluvial cones were formed at the bottom
of the slope and also because sediment overloaded the silt fences on some occasions. For the concave
slope with carbonate colluvium (LCS-CC) the difference between the two values was 47 Mg ha-1,
being 75 % higher the sediment yield estimated from rills. This was likely due to some sediment that
was deposited downslope and did not fill the silt fences.

Vegetation colonization

In our study, the plant establishment pattern was quite different on the linear vs the concave slope
(always regarding topsoiled treatments).

The linear profiles allowed more widespread and homogeneous plant cover. This could be because
their abiotic characteristics: slope angle and surface soil cover depth and compaction which were very
homogeneous, so that its environmental heterogeneity is nor remarkable. In fact, species associated to
worse soil conditions —i.e. Thymus vulgaris, Brachypodium phoenicoides, or Aphyllanthes
monspeliensis— appear only in the linear slope.

The concave profile includes two very different environments (upper steepest part and lower flatter
part). Plant colonization occurred mainly in the lower and flatter one, where water availability as well
as the seed bank richness should be higher. In addition, woody species have been identified here
(Genista scorpius and Sideritis hirsute).

This article is protected by copyright. All rights reserved.

These facts are interpreted as the development of a more ‘structured’ plant community in the concave
slope than in the linear one. In turn, we consider this as a result of a more heterogeneous environment
on the concave slope. Of course, given the very few years of vegetation colonization, these are
preliminary results, and a larger time-span is needed for more conclusive results, as far as the
vegetation development is concerned.

The greater amount of continuous vegetation cover on the linear slope could be another explanation
for the lower sediment yield rates for linear vs concave slopes. In this respect, the value of 50 % of
vegetation cover reached by this linear slope with topsoil and the decrease of sediment yield amount
seems to be in agreement with the literature. Indeed, the role of vegetation cover in sediment yield
control is well known. Several authors have observed that, in mediterranean environments, erosion
rates are greatly reduced when vegetation cover rises up above 30% (Thornes, 2004; de Luís et al.,
2001; Gimeno-García et al., 2007). Andres & Jorba (2000) and Moreno-de las Heras et al. (2009)
confirmed empirically the drastic reduction of soil loss with a 30% plant cover for slopes constructed
for mining reclamation in central and northeast Spain. They recommend a 50% plant cover in practice
as a conservative target. For man-made slopes there is considerable evidence that the restoration of
50% cover with herbaceous vegetation is decisive for site stabilization. And this is what our
experiment seems to show. The literature reflects, however, that it is not only a question of cover, but
also a matter of how the vegetation cover is distributed, such as in natural ecosystems (Cerdá et al.,
2010).

CONCLUSIONS

These conclusions are addressed for mining scenarios similar to the one described, active mines which
already have terraced landforms, with possibility of being improved either by limited topographic
modifications (concave slopes) or by different use of surface soil covers. However, the long term
instability of terraced spoil heaps has been proved, with special emphasis in arid and semi-arid
climates, as the mediterranean one (see Introduction for references). Therefore, wherever mining
reclamation is less conditioned by previous mining works, we recommend a mining reclamation based
in a geomorphic approach, instead of in terraced slopes.

The effect of topography (linear or concave) on soil erosion was prominent when slopes were covered
by carbonate colluvium. Without topsoil, concave slopes yielded much less sediment than linear
slopes, with deposition occurring primarily at the flatter bottom part of the slope, reducing off-site
sediment exportation. Therefore, building concave topographies could be considered advisable when
no topsoil is available.

This article is protected by copyright. All rights reserved.

The interaction between vegetation establishment and topography is complex. Natural plant cover was
more widespread and more homogeneous on linear slopes than on concave ones. In the latter, natural
plant colonization on the steepest part of the concavity was severely limited. The bottom of the
concavity provided more favorable conditions for plant growth.

The three main activities involved in mining reclamation (slope construction, use of surface soil
cover, and plant establishment) did not operate independently in reducing sediment yield and erosion.
This study suggests that the debate about the management of topography and surface soil cover, and
their relationship with vegetation, should be approached under a systemic perspective. The main
trade-offs between major variables should be considered: i) topsoil can lead to a successful mining
reclamation regardless of the two types of topography considered in our experiment; ii) managing
topography by creating concave slopes can lead to a successful mining reclamation when the use of
topsoil is limited; and iii) topsoil and topography determine the plant colonization pattern.

ACKNOWLEDGMENTS

The experiment was funded by a research contract between the Spanish mining company CAOBAR
S.A. and the Department of Geodynamics of the Complutense University of Madrid (research contract
numbers 234/2007, 290/2008 261/2009). The data analyses and manuscript production were
developed within two Research Projects, CGL2009-14508-C02-01 and CGL2010-21754-C02, of the
Spanish Ministry of Science and Technology and by the Ecological Restoration network
REMEDINAL-2 (S2009/AMB-1783).

The authors want to thank several people for their invaluable help: (1) Jonathan B. Laronne for an
earlier review of the manuscript; (2) Lucía Gálvez and Marie Godfrey for help with the translation of
Spanish into English; (3) Ana Lucía Vela and Nacho Zapico for their help at different phases of field
and laboratory work; (4) the Managers of the Upper Tagus Natural Park (Angel Vela, Raquel Ibáñez,
Rafa Ruiz and José Antonio Lozano), for their constant support in improving mining reclamation in
the surrounding areas of the UTNP; (5) Paloma Cubas, of the Department of Plant Biology II of the
Complutense University of Madrid, for her help on field work; and (6) the mining operator company,
Félix Moya S.L., for their involvement in both the experimental spoil heap construction and its
monitoring; (7) the staff rangers of the UPNT who also helped with field work. Finally, two
anonymous reviewers and the editor of the Journal have really helped to improve the final content of
this paper.

This article is protected by copyright. All rights reserved.

 REFERENCES

AEMET. 2012. Agencia Estatal de Meteorología. (Spanish Meteorological Agency).

http://www.aemet.es/es/portada (last accessed Dec. 11, 2012).
Andres P, Jorba M. 2000. Mitigation strategies in some motorway embankments (Catalonia, Spain). Restoration
Ecology 8: 268–275. DOI: 10.1046/j.1526-100x.2000.80038.x.
Barber RG, Romero D. 1994. Effects of bulldozer and chain clearing on soil properties and crop yields. Soil
Science Society of America Journal 58(6): 1768–1775.
Bradshaw AD, Chadwick MJ. 1980. The Restoration of Land. University of California Press: Berkeley.
Brakenhielm S, Liu Q. 1995. Comparison of field methods in vegetation monitoring. Water, Air, and Soil
Pollution 79: 75–87.
Brice JC. 1966. Erosion and deposition in the loess-mantled Great Plains, Medicine Creek drainage basin,
Nebraska. U.S. Geological Survey. Professional Paper Nº 352.
Bugosh N. 2004. Computerizing the fluvial geomorphic approach to land reclamation. In: 2004 National
Meeting of the American Society of Mining and Reclamation and The 25th West Virginia Surface Mine
Drainage Task Force, April 18–24, 2004. Barnhisel RI (ed.). ASMR: Lexington, KY 240–258.
Bugosh N. 2006. Regional variations in stable landforms. And How Critical Elements Can Be Used To Design
Reclamation Landforms. In: Billings Land Reclamation Symposium, June 4-8, 2006, Billings MT. Barnhisel
RI (ed.). BLRS and ASMR: Lexington, KY 156–158.
Carcavilla L, Ruiz R, Rodríguez E. 2008. Guía geológica del Parque Natural del Alto Tajo. Consejería de
Medio Ambiente y Desarrollo Rural. Junta de Comunidades de Castilla-La Mancha. Available at:
http://www.igme.es/internet/patrimonio/descargas.htm (last accessed Dec. 12, 2012).
CEDEX. 2004. CHAC, Cálculo Hidrometeorológico de Aportaciones y Crecidas. Centro de Estudios
Hidrográficos, CEDEX. Ministerios de Fomento y de Medio ambiente. Available at:
http://hercules.cedex.es/Chac/ (last accessed Feb. 6, 2012).
CNIG. 2004. Atlas Nacional de España. Sección II, Grupo 9. Climatología (2ª Edición). Ministerio de Fomento.
Available at: http://www2.ign.es/ane/ane1986-2008/ (last accessed Dec. 12, 2012).
Cerdà A, García-Fayos P. 1997. The influence of slope angle on sediment, water and seed losses on badland
landscapes. Geomorphology 18: 77–90.
Cerdà A, Hooke J, Romero-Diaz A, Montanarella L, Lavee H. (eds.) 2010. Soil erosion on mediterranean type-
ecosystems. Land Degradation and Development, Special Issue. DOI 10.1002/ldr.968.
Chong SK, Cowsert PT. 1997. Infiltration in reclaimed mined land ameliorated with deep tillage treatments. Soil
and Tillage Research 44: 255–264.
Darboux F, Gascuel-Odoux C, Davy P. 2002. Effects of surface water storage by soil roughness on overland-
flow generation. Earth Surface Processes and Landforms 27: 223–233.
Davis Instruments. 2005. Rain Collector II, Product Number: 7852. Rev B. Manual (10/21/05). Available at:
http://www.davisnet.com/weather/products/weather_product.asp?pnum=07852 (last accessed Dec. 12,
2012).

This article is protected by copyright. All rights reserved.

de Luís M, García-Cano MF, Cortina J, Raventós J, González-Hidalgo JC, Sánchez JR. 2001. Climatic trends,
disturbances and short-term vegetation dynamics in a mediterranean shrubland. Forest Ecology and
Management 147: 25–37. DOI: http://dx.doi.org/10.1016/S0378-1127(00)00438-2.
DOCM. 2000. Law 1/2000, 6th april, Diario Oficial de Castilla-La Mancha DOCM 43: 4413-4424.
D'Souza VPC, Morgan RPC. 1976. A laboratory study of the effect of slope steepness and curvature on soil
erosion. Journal of Agricultural Engineering Research 21(1): 21–31. DOI:10.1016/0021-8634(76)90095-0.
Espigares T, Moreno-de las Heras M, Nicolau JM. 2011. Performance of Vegetation in Reclaimed Slopes
Affected by Soil Erosion. Restoration Ecology 19(1): 35–44. DOI: 10.1111/j.1526-100X.2009.00546.x.
Evans KG. 2000. Methods for assessing mine site rehabilitation design for erosion impact. Australian Journey
of Soil Research 38(2): 231–247. DOI:10.1071/SR99036.
Evans MJ, Willgoose GR. 2000. Post-mining evolution landform modelling: 2. Effects of vegetation and surface
ripping. Earth Surface Processes and Landforms 25(8): 803–823. DOI:10.1002/1096-
9837(200008)25:8<803::AID-ESP96>3.0.CO;2-4.
Gimeno-García E, Andreu V, Rubio JL. 2007. Influence of vegetation recovery on water erosion at short and
medium-term after experimental fires in a mediterranean shrubland. Catena 69: 150–160. DOI:
http://dx.doi.org/10.1016/j.catena.2006.05.003.
Gómez JA, Nearing MA. 2005. Runoff and sediment losses from rough and smooth soil surfaces in a laboratory
experiment. Catena 59: 253–266.
González Amuchastegui MJ. 1993. Geomorfología del Alto Tajo en el sector de Molina de Aragón. PhD
Dissertation. Universidad Autónoma, Madrid.
Haigh MJ. 1980. Slope retreat and gullying on revegetated surface mine dumps, Waun Hoscyn, Gwent. Earth
Surface Processes and Landforms 5(1): 77–79. DOI:10.1002/esp.3760050108.
Haigh MJ. 1985. The experimental examination of hill-slope evolution and the reclamation of land disturbed by
coal mining. In: Geography applied to practical problems, Johnson JH (ed.). Geo Books: Norwich 123–138.
Haigh MJ, Sansom, B. 1999. Soil compaction, runoff and erosion on reclaimed coal-lands (UK). International
Journal of Surface Mining, Reclamation and Environment 13(4): 135–146. DOI:
10.1080/09208119908944239.
Haigh MJ. 2000. Erosion Control: Principles and Some Technical Options. In: Reclaimed Land, Erosion
Control, Soils and Ecology, Haigh MJ (ed.). Balkema: Rotterdam 75–110.
Hancock GR, Loch RJ, Willgoose GR. 2003. The design of post-mining landscapes using geomorphic
principles. Earth Surface Processes and Landforms 28(10): 1097–1110. DOI:10.1002/esp.518.
ICONA. 1988. Agresividad de la lluvia en España. MAPA, Madrid.
IUSS Working Group WRB. 2007. World Reference Base for Soil Resources 2006, first update 2007. World
Soil Resources Reports No. 103. FAO, Rome.
Kapolka NM, Dollhopf DJ. 2001. Effect of slope gradient and plant growth on soil loss on reconstructed steep
slopes. International Journal of Surface Mining, Reclamation and Environment 15(2): 86–99. DOI:
10.1076/ijsm.15.2.86.3416.
Köppen W. 1918. Klassifikation der Klimate nach Temperatur, Niederschlag und Jahreslauf. Petermanns Mitt
64: 193–203.

This article is protected by copyright. All rights reserved.

Liu BY, Nearing MA, Shi PJ, Jia ZW. 2000. Slope length effects on soil loss for steep slopes. Soil Science
Society of America Journal 64: 1759–1763.
Loch RJ. 1997. Landform design – better outcomes and reduced costs applying science to above-and below-
ground issues. In: Proceedings of the 22nd Annual Environmental Workshop. Minerals Council of Australia:
Adelaide 550–563.
MARM. 1997–2006. Mapa de Forestal de España Escala 1:50.000, Hoja 433 Atienza, Edición digital.
Ministerio de Medio Ambiente, Medio Rural y Marino, Madrid. Available at:
http://www.marm.es/es/biodiversidad/temas/montes-y-politica-forestal/mapa-forestal/digital_mfe 50.aspx
(last accessed Dec. 12, 2012).
Martín-Duque JF, Sanz MA, Bodoque JM, Lucía A, Martín-Moreno C. 2010. Restoring earth surface processes
through landform design. A 13-year monitoring of a geomorphic reclamation model for quarries on slopes.
Earth Surface Processes and Landforms 35: 531–548. DOI: 10.1002/esp.1950.
Meyer LD, Kramer LA. 1969. Erosion equations predict land slope development. American Society of
Agricultural Engineers 50(9): 522–523.
Moliere DR, Evans KG, Willgoose GR, Saynor MJ. 2002. Temporal trends in erosion and hydrology from a
post-mining landform at Ranger Mine, Northern Territory. Supervising Scientist Report 165. Supervising
Scientist, Darwin NT, Australia.
Moreno-de las Heras M, Merino-Martín L, Nicolau JM. 2009. Effect of vegetation cover on the hydrology of
reclaimed mining soils under mediterranean–continental climate. Catena 77(1): 39–47.
DOI:10.1016/j.catena.2008.12.005.
Moreno-de las Heras M, Nicolau JM, Merino-Martín L, Wilcox BP. 2010. Plot-scale effects on runoff and
erosion along a slope degradation gradient. Water Resources Research 46: W04503.
DOI:10.1029/2009WR007875.
Moreno-de las Heras M, Espigares T, Merino-Martín L, Nicolau JM. 2011. Water-related ecological impacts of
rill erosion processes in mediterranean-dry reclaimed slopes. Catena 84: 114–124.
Morgan RPC. 2005. Soil Erosion and Conservation, 3rd ed. Longmann Group: London, UK.
Nicolau JM, Asensio E. 2000. Rainfall erosion on opencast coal-mine lands: ecological perspective in
Reclaimed Land: Erosion Control, Soils and Ecology, Haigh MJ (ed.). Balkema: Rotterdam 1: 51–73.
Nicolau JM. 2002. Runoff generation and routing on artificial slopes in a mediterranean – continental
environment: the Teruel coalfield, Spain. Hydrological Processes 16: 631–647. DOI:10.1002/hyp.308.
Nicolau JM. 2003. Trends in relief design and construction in opencast mining reclamation. Land Degradation
and Development 14(2): 215–226. DOI:10.1002/ldr.548.
Olmo P, Álvaro M. 1989. Mapa Geológico de España, Escala 1:50.000. 2ª Serie (MAGNA), Peralejos de las
Truchas (539). IGME, Servicio de Publicaciones: Madrid.
Osterkamp WR, Morton RA. 1996. Environmental impacts of urbanization and mining: an international project
on global change. GSA Today 6(7): 14–15.
Pimentel DC, Harvey P, Resosudarmo K, Sinclair D, McNair KM, Crist S, Shpritz L, Fitton L, Saffouri R, Blair
R. 1995. Environmental and economic costs of soil erosion and conservation benefits. Science 267(5201):
1117–1123.

This article is protected by copyright. All rights reserved.

Porta J, Poch R, Boixadera J. 1989. Land evaluation and erosion practices on mined soils in NE Spain. Soil
Technology Series 1: 189–206.
Power JF, Sandoval FM, Ries RE, Merrill SD. 1981. Effects of topsoil and subsoil thickness on soil water
content and crop production on a disturbed soil. Soil Science Society of America Journal 45: 124–129.
Priyashantha S, Ayres B, O’Kane M, Fawcett M. 2009. Assessment of concave and linear hillslopes for post-
mining landscapes. Paper presented at Securing the Future and 8th ICARD, June 23–26, 2009 Skellefteå,
Sweden. Available at: http://www.okc-sk.com/wp-content/uploads/2012/02/Sumith_
Priyashantha_B3_T5_Assessment-of-Concave-and-Linear-Hillslopes-for-Post-Mining-Landscapes. pdf (last
accessed Feb. 6, 2012).
Quansah C. 1981. The effect of soil type, slope, rain intensity and their interactions on splash detachment and
transport. European Journal of Soil Science 32: 215–224. DOI: 10.1111/j.1365-2389.1981.tb01701.x.
Ramos-Scharrón CE, MacDonald LH. 2007. Runoff and suspended sediment yields from an unpaved road
segment, St John, US Virgin Islands. Hydrological Processes 21: 35–50. DOI:10.1002/hyp.6175.
Riley SJ. 1995. Geomorphic estimates of the stability of a uranium mill tailings containment cover, Nabarlek,
NT, Australia. Land Degradation and Rehabilitation 6: 1–16. DOI:10.1002/ldr.3400060102.
Rivas V, Cendrero A, Hurtado M, Cabral M, Gimenez J, Forte L, Del Rio L, Cantu M, Becker A. 2006.
Geomorphic consequences of urban development and mining activities; An analysis of study areas in Spain
and Argentina. Geomorphology 73(3–4): 185-206. DOI:10.1016/j.geomorph.2005.08.006.
Robichaud PR, Brown PR. 2002. Silt fences: An economical technique for measuring hillslope erosion. General
Technical Report RMRS-GTR-94. Rocky Mountain Research Station, USDA Forest Service.
Sanz MA, Martín-Duque JF, Martín-Moreno C, Lucía A, Nicolau JM, Pedraza J, Sánchez L, Ruiz R, García A.
2008. Silica sand slope gullying and mining in Central Spain: erosion processes and geomorphic reclamation
of contour mining. In: Geo-Environment and Landscape Evolution III, Mander U, Brebbia CA, Martín
Duque JF (eds.). Wessex Institute of Technology (WIT Press): Southampton 3–14.
Sawatsky L, Beckstead G. 1996. Geomorphic approach for design of sustainable drainage systems for mineland
reclamation. International Journal of Mining, Reclamation and Environment 10(3): 127–129.
DOI:10.1080/09208119608964815.
Sawatsky LF, Cooper DL, McRoberts E, Ferguson H. 1996. Strategies for reclamation of tailings
impoundments. International Journal of Mining, Reclamation and Environment 10(3): 131-134.
DOI:10.1080/09208119608964816.
Sawatsky L, McKenna G, Keys MJ, Long D. 2000. Towards minimising the long-term liability of reclaimed
mined sites. In: Reclaimed Land: Erosion Control, Soils and Ecology, Haigh MJ (ed.). Balkema: Rotterdam
21–36.
Schor HJ, Gray DH. 2007. Landforming. An environmental approach to hillside development, mine reclamation
and watershed restoration. John Wiley and Sons: Hoboken.
Sobek AA, Shuller WA, Freeman JR, Smith RM. 1978. Field and laboratory methods applicable to overburdens
and mine soils. Environmental Technology Series. EPA–600/2–78–054.
Soulliere EJ, Toy TJ. 1986. Rilling of hillslopes reclaimed before 1977 surface mining law, Dave Johnston
Mine, Wyoming. Earth Surface Processes and Landforms 11: 293–305.

This article is protected by copyright. All rights reserved.

StatPoint Technologies, Inc. 2012 http://www.statlets.com/statgraphics_centurion.htm (last accessed Dec. 12,
2012).
Stefano CD, Ferro V, Porto P, Tusa G. 2000, Slope curvature influence on soil erosion and deposition processes.
Water Resources Research 36(2): 607–617. DOI:10.1029/1999WR900157.
Thornes JB. 2004. Stability and instability in the management of mediterranean desertification. In:
Environmental Modelling: Finding Simplicity in Complexity, Wainwright J, Mulligan M (eds.). Wiley:
Chichester 303–315.
Toy TJ, Hadley RF. 1987. Geomorphology of Disturbed Lands. Academic Press: London.
Toy TJ, Foster GR (eds.). 1998. Guidelines for the Use of the Revised Universal Soil Loss Equation on Mined
Lands, Construction Sites, and Reclaimed Lands. Office of Surface Mining, Reclamation and Enforcement:
Denver.
Toy TJ, Black JP. 2001. Topographic reconstruction: the theory and practice. In: Reclamation of Drastically
Disturbed Lands, Barnishel R, Darmody R, Daniels W (eds.). American Society of Agronomy: Madison 41–
75.
Toy TJ, Foster GR, Renard KG. 2002. Soil Erosion: Processes, Prediction, Measurement and Control. John
Wiley and Sons: New York.
Toy TJ, Chuse W. 2005. Topographic Reconstruction: A Geomorphic Approach. Ecological Engineering 24(1–
2): 29-35. DOI:10.1016/j.ecoleng.2004.12.014.
Vanha-Majamaa I, Salemaa M, Tuominen S, Mikkola K. 2000. Digitized photographs in vegetation analysis – a
comparison of cover estimates. Applied Vegetation Science 3: 89–94.
Whisenant SG, Thurow TL, Maranz SJ. 1995. Initiating autogenic restoration on shallow semiarid sites.
Restoration Ecology 3: 61–67. DOI: 10.1111/j.1526-100X.1995.tb00076.x.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.
Figure 1 Location of the study area within the Iberian Peninsula and within the province of
Guadalajara. The experimental spoil heap is located at El Machorro mine.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.
 Figure 2 Experimental spoil heap of El Machorro mine, during the second study year, after
conversion of the short concave slopes to long concave slopes. Top, treatment scheme;
bottom, photograph taken October 2011, one year after experiment finished. LCS-TS = long
concave slope with topsoil, LCS-CC = long concave slope with carbonate colluvium, LS-CC
= linear slope with carbonate colluvium, LS-TS = linear slope with topsoil, LS-OM = linear
slope with overburden material. The long concave slope with overburden material (LCS-OM)
could not be constructed.

This article is protected by copyright. All rights reserved.

 Figure 3 Concave slope shapes and their n values. The n value is an exponent that varies
according to slope shape, following the equation of Stefano et al. (2000). The original, short
concave slopes were converted to long concave slopes at the end of the first year.

This article is protected by copyright. All rights reserved.

Figure 4 Mean annual sediment yield (Mg ha-1) by treatment and study year. The error bars
represent the standard deviation. SCS-TS = short concave slope with topsoil, LCS-TS =
long concave slope with topsoil, LCS-CC = long concave slope with carbonate colluvium,
LS-CC = linear slope with carbonate colluvium, LS-TS = linear slope with topsoil, LS-
OM = linear slope with overburden material. The short concave slope with carbonate
colluvium (SCS-CC) was not monitored during the first year.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.
Figure 5 Photographs showing geomorphic evolution and vegetation colonization at the
experimental spoil heap (see text for explanation).

This article is protected by copyright. All rights reserved.

Figure 6 Comparison of sediment yield measured from silt fences with sediment yield
estimated from rill erosion, for the two year study period. LCS-CC = long concave slope
with carbonate colluvium, LS-CC = linear slope with carbonate colluvium, LS-OM =
linear slope with overburden material.

This article is protected by copyright. All rights reserved.

 Table I Slope code and starting month and year of measurements for each treatment.
Measurements did not start in October 2008 because the spoil heap was built that month
Treatment
Calendar
Code Month
year
Topographic profile Surface soil cover

SCS-TS Short concave slope topsoil November 2008

LCS-TS Long concave slope topsoil October 2009

LCS-CC Long concave slope carbonate colluvium October 2009

LS-TS Linear slope topsoil November 2008

November 2008
LS-CC Linear slope carbonate colluvium
October 2009
November 2008
LS-OM Linear slope overburden material
October 2009

This article is protected by copyright. All rights reserved.

 Table II Experimental treatments and their characteristics

Surface Slope Slope gradient

Open Area (m2)
Topographic Surface soil cover length (m) (º)
Treatment plot
profile soil cover thickness
number
(cm) Yr 1 Yr 2 Yr 1 Yr 2 Yr 1 Yr 2

1 83.7 91
SCS-TS and
2 topsoil 82.9 104
LCS-TS
3 73.2 100
concave 20 to 30 33 40 4 to 26 4 to 32
4 58.7 106
SCS-CC and carbonate
5 70.3 124
LCS-CC colluvium
6 61.5 100
7 30.9
LS-TS 8 topsoil 35.5

9 45.7

10 27.5
carbonate
LS-CC 11 linear 30 to 75 11 32 23.5
colluvium
12 43.2

13 31.3

LS-OM 14 overburden 43.5

15 31.3

This article is protected by copyright. All rights reserved.

Table III Surface soil cover characteristics and vegetation cover. Values are means.
Vegetation survey was carried out in May 2010

Sand Silt (%)

Clay (%) Organic Bulk Textural Rock
(%) 0.05 – Soil Vegetation
Treatment <0.002 matter density classification cover
2–0.05 0.002 structure cover (%)
mm (%) (g cm-3) (USDA) (%)
mm mm

medium
sandy clay or coarse
LCS-TS 49.7 29.8 20.5 2.3 1.06 20 30
loam granular
2–5mm

medium
or coarse
LCS-CC 39.8 47.2 13.1 0.6 1.26 loam 40 0
granular
2–5mm

fine
LS-TS 39.2 40.8 20.0 3.3 1.09 loam granular 20 50
1–2mm

fine
LS-CC 51.1 36.9 12.1 0.6 1.27 loam granular 25 0
1–2mm

fine
10 to
LS-OM 68.4 16.1 15.5 0.2 1.41 sandy-loam granular 0
5
1–2mm

This article is protected by copyright. All rights reserved.

 Table IV Climate characteristics of each study year
Year First year Second year

Annual rainfall (mm) 434 992

Maximum rainfall (month/mm) Dec 08/125 Dec 09/290

Minimum rainfall (month/mm) Jul 09/1.00 Aug 10/4.20
Max. rainfall in 24h (mm) 38.4 49.0
Average annual temperature (ºC) 10.1 10.3
Maximum average temperature (month/ºC) Aug 09/21.0 Jul 10/20.5
Minimum average temperature (month/ºC) Dec 08/2.00 Jan 10/1.60
Frost free days per year 223 267

This article is protected by copyright. All rights reserved.

 Table V Rainfall characteristics and sediment yield on sampling dates. Total values are also
included by hydrologic year. S/LCS-TS = short/long concave slope with topsoil, LCS-CC =
long concave slope with carbonate colluvium, LS-CC = linear slope with carbonate
colluvium, LS-TS = linear slope with topsoil, LS-OM = linear slope with overburden material

Max. Mean sediment yield (Mg ha-1)/Standard deviation (SD)

# Total
rainfall
Sampling date rain rainfall
24h
days (mm) S/LCS-TS LCS-CC LS-TS LS-CC LS-OM
(mm)
01Oct2008 -
6Nov2008
9 82.6 38.4 Open plots were not yet installed
19Dec2008 16 138 19.2 0.003 /0.003 0.02 /0.03 0.00 /0.00 0.00 /0.00 0.00 /0.00
23Jan2009 14 17.4 4.80 0.00 /0.00 0.00 /0.00 0.00 /0.00 6.38 /9.02 3.27 /4.62
30Jan2009 3 10.6 6.20 0.00 /0.00 - - 0.00 /0.00 29.5 /41.7 44.2 /62.6
12Feb2009 6 11.6 7.60 0.00 /0.00 - - 10.0 /14.2 60.4 /40.3 102 /42.2
13Mar2009 6 3.20 1.00 0.00 /0.00 - - 0.39 /0.29 3.48 /4.14 14.9 /12.3
21Apr2009 16 75.6 12.6 0.00 /0.00 - - 0.32 /0.29 0.78 /0.46 35.4 /45.8
09Jun2009 14 52.8 28.8 1.25 /1.23 - - 0.31 /0.03 14.7 /2.81 43.3 /4.80
24Jun2009 6 6.40 3.60 0.00 /0.00 - - 0.10 /0.05 0.14 /0.02 0.49 /0.14
12Aug2009 6 8.20 6.40 1.30 /0.89 - - 0.34 /0.13 4.90 /1.47 30.0 /2.18
01Oct2009 16 27.6 10.6 0.04 /0.03 - - 0.07 /0.03 0.12 /0.01 8.56 /0.83
1st year total 112 434 - 3 - 12 120 282
Mean 10.2 39.5 12.7 0.26 - 1.16 12.0 28.2
Median 9.00 17.4 7.60 0.00 - 0.20 4.19 22.4
SD 5.04 43.2 11.6 0.54 - 3.12 19.3 31.1
07Oct2009 3 5.80 4.00 0.03 /0.02 0.00 /0.00 0.00 /0.00 0.23 /0.12 3.13 /0.47
29Oct2009 8 43.0 25.0 0.47 /0.44 0.00 /0.00 0.14 /0.10 1.20 /0.56 15.2 /11.8
12Nov2009 9 6.80 2.80 0.00 /0.00 0.00 /0.00 0.00 /0.00 0.00 /0.00 0.20 /0.05
10Dec2009 13 51.6 20.2 0.50 /0.43 0.00 /0.00 0.00 /0.00 0.73 /0.12 19.4 /16.0
18Jan2010 29 328 49.0 8.39 /1.56 7.38 /5.62 0.61 /0.43 23.8 /17.4 102 /12.5
02Mar2010 30 153 24.8 1.29 /0.80 2.35 /1.97 1.67 /1.15 74.6 /34.0 38.1 /7.18
05Apr2010 25 79.4 24.4 0.18 /0.15 1.82 /2.19 0.07 /0.08 2.56 /2.48 16.4 /2.27
19May2010 27 156 26.0 2.04 /1.05 1.97 /2.81 0.09 /0.13 1.44 /0.83 43.0 /16.4
01Jul2010 23 56.8 20.4 2.10 /1.18 2.27 /3.93 0.18 /0.19 5.50 /1.54 67.8 /30.5
30Sep2010 28 35.8 4.20 3.28 /1.06 0.00 /0.00 0.32 /0.11 14.1 /2.94 29.7 /11.2
03Nov2010 17 76.6 22.4 1.90 /0.77 0.00 /0.00 0.00 /0.00 1.42 /0.66 11.8 /3.36
2nd year total 212 992 - 20 16 3 126 347
Mean 19.3 90.2 20.3 1.84 1.44 0.28 11.4 31.6
Median 23.0 56.8 22.4 1.29 0.00 0.09 1.44 19.4
SD 9.67 93.2 13.2 2.42 2.23 0.50 22.2 30.5

2 year total 324 1426 - 23 16 15 246 629

This article is protected by copyright. All rights reserved.

Table VI Results of paired t-test. Statistical significance level: *α=0.05. S/LCS-TS =
short/long concave slope with topsoil, LCS-CC = long concave slope with carbonate
colluvium, LS-CC = linear slope with carbonate colluvium, LS-TS = linear slope with
topsoil, LS-OM = linear slope with overburden material

Study Treatments T-test results

year compared P value

SCS-TS vs LS-TS 0.38

LS-TS vs LS-CC 0.09
2009

LS-TS vs LS-OM 0.01*

LS-CC vs LS-OM 0.18
LCS-TS vs LCS-CC 0.69
LCS-TS vs LS-TS 0.05*
LCS-CC vs LS-CC 0.15
2010

LS-TS vs LS-CC 0.11

LS-TS vs LS-OM <0.01*
LS-CC vs LS-OM 0.09
2009+2010

LS-TS vs LS-CC 0.02*

LS-TS vs LS-OM <0.01*
LS-CC vs LS-OM 0.03*

This article is protected by copyright. All rights reserved.

Table VII Plant species established in the slopes with topsoil.

Concave slope with topsoil Linear slope with topsoil

Family Compositae
Cuprina crupinastrum Hieracium pilosella
Leucanthemum vulgare
Family Euphorbiaceae
Euphorbia sp. Euphorbia sp.
Family Gramineae (=Poaceae)
Arrhenatherum elatius subsp.
Brachypodium phoenicoides
bulbosum
Festuca gr. rubra Bromus erectus
Family Lamiaceae
Sideritis hirsuta Thymus vulgaris
Family Leguminosae (=Fabaceae)
Coronilla repanda Coronilla repanda
Genista scorpius Lotus corniculatus
Medicago lupulina Medicago lupulina
Family Liliaceae
Aphyllanthes monspeliensis
Family Plantaginaceae
Plantago sp.
Family Rosaceae
Filipendula vulgaris Rosa sp.

Sanguisorba minor Sanguisorba minor

Family Rubiaceae
Asperula montana
Galium lucidum Galium lucidum
Family Resedaceae
Reseda alba
Reseda phyteuma

This article is protected by copyright. All rights reserved.

 Table VIII Concave and linear profile characteristics related to sediment yield, rill
development, and establishment of vegetation

Topographic Sediment yield Soil surface

Runoff control Natural plant colonization
profile control cover

Watershed size and runoff increase heterogeneous distribution

Sediment ↑ compaction
downslope, while slope gradient plant colonization more
accumulates at ↓ thickness
decreases. difficult in steep upper part
Concave slope lower, flat part heterogeneous
Decrease of energy downslope. of slope than in the lower
of slope distribution
part

Watershed size and runoff increase ↓ compaction

downslope, while slope gradient is ↑ thickness
None homogeneous distribution
Linear slope constant. homogeneous
Increase of energy downslope. distribution

This article is protected by copyright. All rights reserved.