PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Badía, David]On: 1 October 2008Access details: Access Details: [subscription number 903255980]Publisher Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Arid Land Research and ManagementPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713926000

Erodibility and Hydrology of Arid Burned Soils: Soil Type and RevegetationEffectsDavid Badía a; Clara Martí a; Javier Aguirre a; Maria Teresa Echeverría b; Paloma Ibarra b

a High School of Agricultural Engineering, University of Zaragoza, Huesca, Spain b Department of Geographyand Territorial Management, University of Zaragoza, Zaragoza, Spain

Online Publication Date: 01 October 2008

To cite this Article Badía, David, Martí, Clara, Aguirre, Javier, Echeverría, Maria Teresa and Ibarra, Paloma(2008)'Erodibility andHydrology of Arid Burned Soils: Soil Type and Revegetation Effects',Arid Land Research and Management,22:4,286 — 295

To link to this Article: DOI: 10.1080/15324980802388231

URL: http://dx.doi.org/10.1080/15324980802388231

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

Erodibility and Hydrology of Arid Burned Soils: SoilType and Revegetation Effects

David Badıa1, Clara Martı1, Javier Aguirre1, Maria TeresaEcheverrıa2, and Paloma Ibarra2

1High School of Agricultural Engineering, University of Zaragoza,Huesca, Spain.2Department of Geography and Territorial Management, University ofZaragoza, Zaragoza, Spain.

A set of experiments with a rainfall simulator was performed in the field to evaluatethe efficiency of parallel contour seeding as a post-fire restoration strategy in thearid central Ebro Valley (NE Spain). Rainfall simulations were conducted in spring,after seeded plant development, on calcareous and gypsiferous soils, with the sameexperimental design (two-soil treatments—seeded and nonseeded—per two soiltypes and per nine replicates). The parallel contour seeding treatment increases soilcover and soil surface roughness, which significantly ameliorates the hydrologicaland erosional response of both calcareous and gypsiferous burned soils. Seedingdecreased soil loss, both in calcareous (23-fold) and gypsiferous soils (4-fold). Inaddition, it decreased the sediment concentration of runoff for calcareous (6-fold)and gypsiferous soils (2-fold) and the runoff coefficient for calcareous (3.5-fold)and gypsiferous soils (1.5-fold). On the other hand, seeding increased the steadystate infiltration rate (3-fold), as well as surface soil moisture (1.2-fold) andwetting front depth (2-fold), with a similar order of magnitude for both soils. Timeto runoff and runoff quality (electrical conductivity [EC] and pH) were not affec-ted by seeding. Gypsiferous soils had a higher soil loss, runoff coefficient, and ECflow and a lower time to runoff, steady state infiltration rate, and wetting frontdepth than calcareous soils. Some of these differences were directly related to differ-ences in soil gypsum and carbonate content (i.e., EC and pH of water runoff) andothers are related to the proportion of surface soil protection, an indirect effect ofsoil characteristics. The results suggest that parallel contour seeding is an effectiverestoration measure with a short-term response that allows conservation of waterand soil on recently burned arid lands, especially in situations of high erosion riskas occurs with soils with low plant cover.

Keywords calcareous and gypsiferous soils, Ebro Valley, forest fires, infiltration,parallel contour seeding, rainfall simulator, runoff, soil loss

Received 28 December 2007; accepted 16 June 2008.This research was part of the project, ‘‘Space-temporary behavior and models of the

sediment transfer on different land uses: the role of forest fires,’’ supported by the Spanishgovernment: National Plan IþDþ I on Biodiversity, Earth Sciences, and Global Change,Ref. CGL2007-66644-C04-04=HID. The authors express their gratitude for valuablecomments from the editor and anonymous reviewers.

Address correspondence to Dr. David Badıa, High School of Agricultural Engineering,Escuela Politecnica Superior, Crtra. Cuarte s=n, Huesca 22071, Spain. E-mail: badia@unizar.es

Arid Land Research and Management, 22:286–295, 2008Copyright # Taylor & Francis Group, LLCISSN: 1532-4982 print/1532-4990 onlineDOI: 10.1080/15324980802388231

286

Downloaded By: [Badía, David] At: 08:25 1 October 2008

Introduction

Wildfire drastically reduces vegetation and litter cover, modifying soil erodibility andhydrology.These changes often result in enhanced soil erosion and overland flow onslopes (Shakesby and Doerr, 2006). In Mediterranean conditions, where severe wild-fires occur mainly in summer and torrential events are frequent in autumn, short-term conservation measures are necessary to prevent irreversible soil degradationin the first vegetation regeneration phase after a fire (Vallejo and Alloza, 1999;De Luis, Gonz�aalez-Hidalgo, and Ravent�oos, 2003; Rulli, Bozzi, Spada, Bocchiola,and Rosso, 2006). A post-fire measure to improve soil protection is revegetation.Some authors suggest that the vegetation type can be as important as the vegetationcover in preventing soil erosion (Desir, Sirvent, Gutierrez, and Sancho, 1995; Cerdaand Doerr, 2005). In addition, it has been suggested that shrubs and grasses may bemore effective than trees in reducing erosion in arid regions (Quine, Navas, Walling,and Machın, 1994; Cammeraat and Imeson, 1999).

A predicted rise in temperature over the next century would increase wildfireprobability in forested arid lands (Moreno, 2005). Moreover, Johansen, Hakonson,and Breshears (2001) have suggested that forests are more vulnerable than shrub-lands and grasslands to increases in post-fire erosion. On the other hand, therehave been scattered and conflicting results reported on the efficiency of post-fireremediation strategies, such as seeding, in reducing water and soil losses inMediterranean conditions (Vallejo and Alloza, 1999; Marcos, T�aarrega, and Luis-Calabuig 2000). Runoff and soil erosion rates return to undisturbed values whenground vegetation and litter cover the soil surface, although rebound periodscan vary, being as short as months or as long as decades, because of soil typeand aridity (Badıa and Martı, 2008). The variability of these effects depends onan often-complex interplay of factors including climate, vegetation, terrain, andfire characteristics (Cerda and Doerr, 2005). Rainfall simulation experiments allowcomparison of soil erodibility in selected microplots with defined fire history,terrain, rainfall characteristics, and post-fire treatments. Furthermore, although agrowing body of literature has addressed fire-induced hydrological and geomor-phological impacts, there is still little information on arid fire-prone terrain(Shakesby and Doerr, 2006).

The objective of the present research was to determine the effects of parallelcontour seeding, as a measure of post-fire rehabilitation, using two contrasting soils(calcareous soil and gypsiferous soil) of arid lands by means of a rainfall simulatortechnique.

Material and Methods

Terrain

A soil developed on Oligocene marls (calcareous soil), classified as Xeric Torriorthent(Calcaric Regosol), and another developed on Oligocene gypsum (gypsiferous soil),classified as Xeric Haplogypsid (Gypsisol hypergypsic), were studied because of theirwide distribution on the slopes of the semi-arid Ebro Valley (Badıa, Ibarra, Martı,Longares, and Belmonte, 2008). Plots sampled were placed, in N-facing slopes, atan elevation of about 200–300 m. a.s.l., about 41�300 N and 0�180E (Figure 1). The soiltemperature regime was thermic, and the soil moisture was aridic ranging to xeric. The

Revegetation Effects on Soil Erosion Control 287

Downloaded By: [Badía, David] At: 08:25 1 October 2008

native vegetation was burned 1 year before the seeding treatment. At the moment ofseeding treatment, vegetation was scarce, dominated by small resprouts of Ononis tri-dentata L. and seedlings of Herniaria fruticosa L. in gypsiferous soil and byBrachypodium retusum (Pers.) Beauv. in calcareous soil.

Paired seeded and control (nonseeded) plots were established in nine differentmicroplots and for two types of soil (two-soil treatment per two soil types and pernine replicates¼ 36 rainfall simulations). Some landscape and soil properties(0–5 cm) of the experimental microplots (8 m2, included with other microplots, infenced plots of 50 m2) before rainfall experiments are reported in Table 1.

Figure 1. Location of studied area.

288 D. Badıa et al.

Downloaded By: [Badía, David] At: 08:25 1 October 2008

Seeding Treatment

The seeding treatment consisted of an herb seed mixture composed of Medicagosativa L., Medicago truncatula Gaertn., Onobrychis viciifolia Scop., Vicia villosaRoth, Agropyron cristatum L. Gaertn., Dactylis glomerata L., Lolium rigidum Gaud.,and Phalaris canariensis L. We broadcasted the seed mixture in rows,perpendicularly to the maximal slope (parallel contour seeding) to increase soilroughness as well as plant cover. The sowing rate for the whole mixture was30 g m�2, with an equivalent weight for each species. The seed mixture was surfaceapplied by hand in autumn, 1 year after a wildfire, and the rainfall simulations werecarried out in the following spring when plants had grown (Figure 2).

Figure 2. Detail of control and seeded microplots.

Table 1. Some landscape and soil properties (0–5 cm) of the experimental plots in thesemi-arid Central Ebro Valley (mean� standard deviation of nine replicates) beforerainfall experiments

Microplot propertiesNon-seeded Seeded

Soil type Calcareous Gypsiferous Calcareous Gypsiferous

Slope (%) 31.9� 5.3 28.4� 6.9 28.9� 6.4 27.8� 6.1Bare soil (%) 66.7� 13.0 70.7� 8.8 27.6� 8.6 50.3� 6.9Surface stoniness (%) 23.9� 9.9 9.5� 5.8 25.5� 14.0 6.1� 4.1Soil moisture (g kg�1) 172� 21 184� 29 166� 24 168� 25pH (1:2.5) 8.3� 0.06 7.9� 0.04 8.2� 0.05 7.9� 0.03Organic matter (g kg�1) 40.5� 4.0 13.9� 3.5 43.0� 3.0 13.2� 1.3Carbonates (g kg�1) 539.0� 41.0 76.7� 34.8 494.9� 39.3 44.2� 14.3Gypsum (g kg�1) <80 427.4� 61.0 <80 533� 62.9Field capacity (g kg�1) 292� 35 385� 71 285� 24 394� 99

Revegetation Effects on Soil Erosion Control 289

Downloaded By: [Badía, David] At: 08:25 1 October 2008

Rainfall Simulation Experiments

Experiments were made using a portable, sprinkler-based rainfall simulator that hasbeen previously described (Badıa and Martı, 2008). A rainfall simulation experimentconsisted of a single, 60-minute rainfall application at a constant intensity on a0.24-m2 surface area. The simulator provided an 85� 8.0 mm h�1 intensity rainfallwith a raindrop size distribution varying between 0.5 and 3.5 mm. Flow and pressurewere guaranteed by a Masterflex pump (Masterflex AG, Gefenkirchen, Germany),which pulled the water from a set of 30-L capacity reservoirs. All simulations weremade with water used to irrigate, with an electrical conductivity of 400�27.5 mS cm�1 (0.4 dS m�1) and a pH of 8.0� 0.1. Lightweight plastic was placedaround each microplot to prevent wind disturbance.

A 5-cm-high sheet of metal placed in the ground isolated each microplot. Surfacerunoff (i.e., overland flow discharge) was delivered to a single exit point situated atthe lowest part of the microplot and manually collected in 1.5-L bottles. When run-off appeared, runoff samples were taken along any single rainfall simulation event atfive different times (40, 80, 120, 440, and 600). Its sediment content was measured in thelaboratory following complete evaporation of the water. Runoff samples were ana-lyzed for determination of the temporal evolution of the pH, electrical conductivity(EC), sediment loss rate (kg ha�1 per mm of rainfall) and sediment concentration(g L�1), runoff (mm h�1), and runoff coefficient (runoff=rainfall ratio, in percent-age). Infiltration (mm h�1) was calculated from the differences between mean rainfalland overland flow measured at defined intervals. Infiltration values were used toobtain steady state infiltration rates according to the Horton model (Horton,1940). Following each simulation, the soil plots were excavated and the depths ofthe continuous wetting fronts recorded (wetting front depth). Soil samples to deter-mine the moisture content were taken before each experiment at the surface (0–5 cm)and after each experiment every 5 cm to the end of the wetting front (only data for0–5 cm have been reported here). The soil water content was measured by gravimetry.

Statistical Analysis

A two-way analysis of variance (ANOVA) was used to evaluate the effects of seedingand soil type on erodibility and hydrology of burned soils. All variables were trans-formed to meet normality criteria and assumptions for parametric statistical analysis(SAS, 1998). Correlations between erosional and hydrological parameters measuredand bare soil, organic matter, and calcium carbonate and gypsum contents were alsodetermined (Pearson rank correlation test).

Results and Discussion

Erodibility Behavior

Both seeding treatment and soil type significantly affected soil loss and sediment concen-tration. Soil erosion values on nonseeded plots were 592.2 g m�2 h�1 (about 70 kg ha�1

mm�1) for calcareous and 659.1 g m�2 h�1 (78 kg ha�1 mm�1) for gypsiferous soil.Seeding treatment decreased soil losses to 25.7 g m�2 h�1 (3 kg ha�1 mm�1)for calcareous and 174.9 g m�2 h�1 (21 kg ha�1 mm�1) for gypsiferous soils. Thisdecrease in soil losses is statistically significant for both soils and related with soil cover(Table 2).

290 D. Badıa et al.

Downloaded By: [Badía, David] At: 08:25 1 October 2008

Ta

ble

2.

Res

ult

so

fth

era

infa

llsi

mu

lati

on

exp

erim

ents

an

do

fth

etw

o-w

ay

AN

OV

Ate

sts

mea

suri

ng

the

effe

cts

of

trea

tmen

ta

nd

soil

typ

eo

nth

eer

osi

on

al

an

dh

yd

rolo

gic

al

resp

on

seo

fth

eb

urn

edso

ils.

Nin

ere

pli

cate

sfo

rea

chtr

eatm

ent

Tre

atm

ents

No

nse

eded

See

ded

Pro

ba

bil

ity

So

ilty

pe

Ca

lca

reo

us

Gy

psi

fero

us

Ca

lca

reo

us

Gy

psi

fero

us

Sta

nd

ard

erro

rm

ean

Tre

atm

ent

So

ilT

rea

tmen

t�

So

ilty

pe

So

illo

ss(g

m�

2h�

1)

59

2.2

65

9.1

25

.71

74

.90

.18

6<

0.0

01

0.0

14

0.2

34

Sed

imen

tco

nce

ntr

ati

on

(gL�

1)

15

.34

13

.42

2.4

75

.48

0.0

95

<0

.00

10

.00

60

.02

6

Infi

ltra

tio

nra

te(m

mh�

1)

21

.41

3.6

63

.94

3.6

0.1

26

<0

.00

1<

0.0

01

0.8

76

Su

rfa

ceso

ilm

ois

ture

(gk

g�

1)a

40

24

47

47

45

29

1.5

29

0.0

02

0.0

27

0.8

10

Wet

tin

gfr

on

td

epth

(cm

)1

3.0

8.0

24

.51

8.0

0.1

84

<0

.00

1<

0.0

01

0.3

90

Tim

eto

run

off

(sec

)1

51

.11

06

.11

71

.11

26

.31

2.1

30

.10

7<

0.0

01

0.9

93

Ru

no

ffco

effi

cien

t(%

)4

5.4

57

.81

2.8

37

.50

.09

5<

0.0

01

<0

.00

10

.59

7E

C(l

Scm�

1)b

47

21

84

94

36

16

90

0.2

70

0.1

01

<0

.00

10

.62

7p

Hb

8.3

77

.91

8.3

67

.90

0.1

45

0.2

07

<0

.00

10

.32

2

aA

fter

rain

fall

sim

ula

tio

n(0

–5

cmd

epth

).b

Av

era

ge

of

fiv

eru

no

ffsa

mp

les,

tak

ena

lon

ga

ny

sin

gle

rain

fall

sim

ula

tio

nex

per

imen

ta

tfi

ve

dif

fere

nt

tim

esd

uri

ng

an

ho

ur.

291

Downloaded By: [Badía, David] At: 08:25 1 October 2008

Seeding treatment decreased bare soil from 70.7% to 50.3% in gypsiferous soiland from 66.7% to 27.6% in calcareous soil where the resprouts of naturalvegetation (such as B. retusum) facilitated these changes. Soil loss was positivelyand significantly correlated (Table 3) to bare soil (r¼ 0.857) demonstrating thatwhen natural plant recovery was very slow, the seeding treatment improved plantcover and reduced erosion in burned soils. Soil loss is higher in gypsiferous thanin calcareous soils, which is related to their proportion of bare soil influenced bysurface stoniness and plant growth. Moreover, gypsiferous soils show a reducedpresence of resprouters, related to a low fertility (Desir et al., 1995; Lasanta, Garcıa,Perez, and Sancho, 2000). In arid conditions, where vegetation cover is scarce, theerosional and hydrological behavior of slopes can be controlled by soil parentmaterial (Vallejo and Alloza, 1999; Cerda, 2002).

Cerda, Imeson, and Calvo (1995) found values of soil loss of about70 kg ha�1 mm�1 in a Mediterranean garrigue or matorral vegetation that hadburned 2 years prior to assessment on the north-facing slopes of Canals (SE Spain).Although erosional and hydrological properties recovered each year after the fire,the original values were not achieved until 6 years following that forest fire (Cerda,1998). In addition, Rodriguez, Vicen, Badıa, and Ascaso (2000) found a progressivedecrease in soil erosion related to plant recovery in burned pine forests of Castej�oonde Valdejasa Hills (Central Ebro Valley, NE Spain). However, torrential rainfallevents 2 years after the wildfire increased the soil erosion 100 times in comparisonto unburned plots (Rodriguez et al., 2000). When almost 100% of soil cover hadbeen recovered in plots burned 11 years previously, negligible erosion rates wereobserved (Cerda and Doerr, 2005). Ortiz and Alca~nniz (2001) sowed a mixture ofDactylis glomerata, Festuca rubra, and Lolium perenne (20 g m�2) on a wet Luvisolin Taradell (NE Spain) where plants covered 100% of the soil in a few weeks afterseeding. They found a soil erosion rate of about 5 kg ha�1 mm�1 in seeded plots, asignificantly lower value than that for burned, bare soil (about 95 kg ha�1 mm�1).

Table 3. R values (Pearson rank correlation test) for erosional and hydrologicalparameters and bare soil, soil organic matter, and carbonate and gypsum content;a0.05¼ 0.325; a0.01¼ 0.418 (n¼ 36)

Parameter Bare soil Organic matter Carbonates Gypsum

Soil loss 0.857 �0.133 �0.044 0.074Sediment concentration 0.567 �0.046 0.081 �0.049Infiltration rate �0.915 0.354 0.210 �0.426Surface soil moisture 0.751 �0.175 �0.276 0.317Wetting front depth �0.732 0.403 0.292 �0.314Time to runoff �0.659 0.447 0.459 �0.540Runoff coefficient 0.777 �0.529 �0.382 0.408EC of runoff 0.871 �0.973 �0.961 0.903pH of runoff �0.603 0.840 0.890 �0.878Soil organic matter �0.811 1 0.934 �0.938Soil carbonate content �0.769 0.934 1 �0.972Soil gypsum content 0.761 �0.938 �0.972 1

292 D. Badıa et al.

Downloaded By: [Badía, David] At: 08:25 1 October 2008

Hydrology Behavior

Seeding significantly increased steady state infiltration rates, soil moisture, andwetting front and decreased runoff coefficients for both soil types. Seeding increasedsoil infiltration in gypsiferous soils from 13.6 to 43.6 mm h�1 and in calcareous soilsfrom 21.4 mm h�1 to 63.9 mm h�1. This result indicated that the seeding treatmentincreased soil infiltration by about three times for both soils (Table 2).

Wetting front depth on nonseeded plots was about 8 to 13 cm, increasing twiceafter seeding. In addition, surface soil moisture was significantly higher in seededplots than nonseeded ones (P¼ 0.002), with differences between soils (P¼ 0.027).

Soil infiltration rates and related parameters, which are modified after a fire,usually recover the original rates with time, when plant cover increases (Cerdaand Doerr, 2005). Cerda (1998) found in a Mediterranean shrubland the loweststeady state infiltration capacity just after a fire (25 mm h�1); values increased gradu-ally during the post-fire recovery period until the sixth year after the fire, whenthe natural infiltration rate was recovered (52 mm h�1). With seeding treatment inparallel contour lines, these values recovered quickly without any actuation.

Time to runoff (lag time) was significantly higher for calcareous soil than forgypsiferous soil. Although there was a delay in the time to runoff in relation to bare soilproportion (r¼ 0.777; P< 0.01), differences by seeding were not significant (P¼ 0.107).

Seeding treatment and soil types significantly changed the accumulated runoff(Figure 3) and thus also the overland flow coefficients. On nonseeded plots, overlandflow coefficients were about 45.4% for calcareous soils and 57.8% for gypsiferoussoils, decreasing significantly with seeding treatment to 12.8% and 37.5%, respect-ively. These values are quite similar to experiments made in summer-dry conditionsand lower than the same soils just after burning (Badıa and Martı, 2008).

Seeding treatment did not affect EC and pH of overland flow, although signifi-cant differences were found between soils, which were associated with the gypsum=calcite content and their solubility. The pH was about 8.4 in calcareous soils and

Figure 3. Evolution of accumulated runoff in different soils (gypsiferous and calcareous) andtreatments (control and seeded).

Revegetation Effects on Soil Erosion Control 293

Downloaded By: [Badía, David] At: 08:25 1 October 2008

7.9 in gypsiferous soils. The EC ranged between 436 and 472 mS cm�1 in calcareoussoils (seeded and nonseeded, respectively), while in gypsiferous soil, EC rangedbetween 1690 and 1849 mS c m�1 (Table 2). The runoff EC was significantly(P< 0.01) correlated with soil gypsum content (r¼ 0.943), as was runoff pH with soilcarbonate content (r¼ 0.890).

Gypsiferous soil had a significantly lower infiltration value, time to runoff, andwetting front depth and higher runoff than calcareous soil (Figure 3). There are vari-ous explanations for these differences. The growth of gypsum crystals in pre-existingpores decreases water flow in gypsiferous soils (Poch and Verplancke, 1997). More-over, gypsiferous soils have low organic matter content and aggregate stability thatelicit a slow micropore flow (Martı, Badıa, and Buesa, 2001). This feature has beenobserved in similar soils where microcrust developed (Lasanta et al., 2000; Desiret al., 1995). Desir (2002), studying natural gypsiferous soils covered by small shrubs,observed that water infiltration occurred only in a few centimeters of the soil, pro-ducing false saturation conditions, and as a consequence, high runoff (Hortonianor infiltration-excess overland flow).

Conclusions

Seeding treatment, as a post-fire restoration strategy for a short-term response, sig-nificantly ameliorates the hydrological and erosional response of both calcareousand gypsiferous soils. Seeding decreased soil loss, both in calcareous (23-fold) andgypsiferous (4-fold) soils. In addition, it decreased sediment concentration of runofffor calcareous (6-fold) and gypsiferous (2-fold) soils and the runoff coefficient forcalcareous (3.5-fold) and gypsiferous soils (1.5-fold). On the other hand, seedingincreased the steady state infiltration rate (3-fold), as well as wetting front depth(2-fold), with a similar order of magnitude for both soils, but did not affect timeto runoff or runoff quality (EC and pH). The recovery of erosional and hydrologicalproperties after fire was more rapid for calcareous than for gypsiferous soils, which isrelated to differences in soil characteristics. Some of the differences arise from soilchemical properties, such as the variation in EC and pH of water runoff with soilgypsum and carbonate content. Other differences are related to soil surface protec-tion, mainly derived from plant cover, an indirect effect of soil properties.

The seeding treatment was thus important in two ways: First, it aided in forma-tion of a vegetation canopy that intercepted rainfall and decreased the final dropimpact velocity, avoiding crust formation. Second, parallel contour seeding resultedin small lines of plants that increased soil surface roughness and acted as a barrier toreduce flow velocity while increasing infiltration and trapping sediments. The seed-ing treatment assayed here improved soil protection on the more erodible gypsifer-ous soils, especially in situations of high erosion risk, and with vegetation thatincludes few resprouter species.

References

Badıa, D. and C. Martı. 2008. Fire and rainfall energy effects on soil erosion and runoffgeneration in semi-arid forested lands. Arid Land Research and Management 22: 93–108.

Badıa, D., P. Ibarra, C. Martı, L. A. Longares, and A. Belmonte. 2008. El Aiguabarrreig: Soilsand landscapes. Ed. CPNA Research Series, Vol. 53, 193 pp. Zaragoza, Spain.

294 D. Badıa et al.

Downloaded By: [Badía, David] At: 08:25 1 October 2008

Cammeraat, L. H. and A. C. Imeson. 1999. The evolution and significance of soil-vegetationpatterns following land abandonment and fire in Spain. Catena 37:107–127.

Cerda, A. 1998. Changes in overland flow and infiltration after a rangeland fire in a Mediter-ranean scrubland. Hydrological Processes 12:1031–1042.

Cerda, A. 2002. The effect of season and parent material on water erosion of highly erodedsoils in eastern Spain. Journal of Arid Environments 52:319–337.

Cerda, A. and S. H. Doerr. 2005. Influence of vegetation recovery on soil hydrology anderodibility following fire: An 11-year investigation. International Journal of Wildland Fire14:423–437.

Cerda, A., A. C. Imeson, and A. Calvo. 1995. Fire and aspect induced differences on theerodibility and hydrology of soils at La Costera, Valencia, southeast Spain. Catena24:289–304.

De Luis, M., J. C. Gonz�aalez-Hidalgo, and J. Ravent�oos. 2003. Effects of fire and torrentialrainfall on erosion in a Mediterranean gorse community. Land Degradation and Develop-ment 14:203–213.

Desir, G. 2002. Hydrological response types for gypsiferous soils in a semiarid region duringnine years of continuous record. Hydrological Processes 16:2685–2700.

Desir, G., J. Sirvent, M. Gutierrez, and C. Sancho. 1995. Sediment yield from gypsiferousdegraded areas in the middle Ebro Basin. Phys. Chem. Earth, 20:385–393.

Horton, R. E. 1940. An approach toward a physical interpretation of infiltration capacity.Proceeding of Soil Science Society of America 5:399–417.

Johansen, M. P., T. E. Hakonson, and D. D. Breshears. 2001. Post-fire runoff and erosionfrom rainfall simulation: contrasting forests with shrublands and grasslands.Hydrological Processes 15:2953–2965.

Lasanta, T., J. M. Garcıa, C. Perez, and C. Sancho. 2000. Runoff and sediment yield in asemiarid environment: the effect of land management after farmland abandonment.Catena 38:265–278.

Marcos, E., R. T�aarrega, and E. Luis-Calabuig. 2000. Comparative analysis of runoff andsediment yield with a rainfall simulator after experimental fire. Arid Soil Research andRehabilitation 14:293–307.

Martı, C., D. Badıa, and M. A. Buesa. 2001. Aggregate stability in Altoarag�oon soils:Comparison of wet steving and lab rainfall simulation methods. Edafologıa 8:21–30.

Moreno, J. M. (Coord.). 2005. Preliminary evaluation of the impacts of climate change in Spain.ECCE. Environmental Ministry, Madrid Project, Spain.

Ortiz, O. and J. M. Alca~nniz, 2001. Erosion reduction of the alfisol due to sewage sludge appli-caion or mineral fertilization to promote vegetation cover. Orsis 16:121–131.

Poch, R. M. and H. Verplancke. 1997. Penetration resistance of gypsiferous horizons.European Journal of Soil Science 48:535–543.

Quine T. A., A. Navas, D. E. Walling, and J. Machın. 1994. Soil erosion and redistribution incultivated and uncultivated land near Las Bardenas in the Cental Ebro River Basin. LandDegradation and Rehabilitation 5:41–55.

Rodriguez, J. M., F. J. Vicen, D., Badıa, and J. Ascaso. 2000. Fire effects on plant recoveryand soil erosion in Castejon de Valdejasa Hills (Zaragoza). Georgica 7:55–68.

Rulli, M. A., S. Bozzi, M. Spada, D. Bocchiola, and R. Rosso. 2006. Rainfall simulations on afire disturbed mediterranean area. Journal of Hydrology 327:323–338.

SAS. 1998. SAS=STAT Software User’s guide. Release 6.12. SAS Institute Inc., NC.Shakesby, R. A. and S. H. Doerr 2006. Wildfire as a hydrological and geomorphological

agent. Earth-Science Reviews 74:269–307.Vallejo, R. and J. A. Alloza. 1999. The restoration of burned lands: The case of eastern Spain,

pp. 91–108, in J. M. Moreno, ed. Large forest fires. Backhuys Publishers, Leiden. TheNetherlands.

Revegetation Effects on Soil Erosion Control 295

Downloaded By: [Badía, David] At: 08:25 1 October 2008