The impact of the farming practice of remodelling hillslope topography on badland morphology and soil erosion processes

Ž .Catena 40 2000 229–250www.elsevier.comrlocatercatena

The impact of the farming practice of remodellinghillslope topography on badland morphology and

soil erosion processes

Michele L. Clarke a,), Helen M. Rendell b`a Centre for EnÕironmental Management, School of Geography, UniÕersity of Nottingham, UniÕersity Park,

Nottingham NG7 2RD, UKb Geography Laboratory, Arts C, UniÕersity of Sussex, Falmer, Brighton BN1 9QN, UK

Received 13 January 1998; received in revised form 11 August 1998; accepted 23 September 1998

Abstract

Badland landforms, created in Plio–Pleistocene clay landscapes of the Basilicata region ofsouthern Italy, form marginal features in a semi-arid landscape dominated by the widespreadcultivation of subsidised cereals. These badland features are high relative relief forms exhibiting ahigh drainage density and steep slopes, with slope angles typically in excess of 358. Economicincentives to increase agricultural productivity have resulted in the practice of remodelling thesemarginal areas using heavy earth-moving equipment. Remodelling the badland features createslonger slopes at lower angles, which can be cultivated using conventional farm machinery. Thesechanges in hillslope morphology have altered the degree to which soil erosion processes operatingin these areas are spatially coupled. In the badland areas, erosion and deposition are stronglylocalised with minimal sediment delivery to ephemeral or perennial channel systems. Theeconomically-driven change in land use from visually striking badland areas to newly remodelledfields for agricultural use results in an increase in the coupling of drainage networks and a netincrease in soil erosion. q 2000 Elsevier Science B.V. All rights reserved.

Keywords: Badlands; Calanchi; Biancane; Soil erosion; Land use change; Common agricultural policy

1. Introduction

Visually striking badland landscapes are found in fine-grained clastic sedimentarybedrock regions of central and southern Italy, principally in the provinces of Marche

) Corresponding author. E-mail: [email protected]

0341-8162r00r$20.00 q 2000 Elsevier Science B.V. All rights reserved.Ž .PII: S0341-8162 99 00047-8

( )M.L. Clarke, H.M. RendellrCatena 40 2000 229–250230

Ž . Ž . ŽCalzecchi-Onesti, 1957 , Tuscany Guasparri, 1993 and Basilicata Rendell, 1982,.1986 . These badlands consist of deeply-dissected, unvegetated or poorly-vegetated

landforms of high relative relief and high drainage density. In the region of BasilicataŽ .formerly known as Lucania it has been estimated that badlands occupy 32,125 ha in

Ž .the Bradano, Basento, Cavone and Agri catchments Radina, 1964 . The badland areasare located in the middle and lower reaches of these catchments on Plio–Pleistocene

Ž .marine clays, which were deposited as part of the Fossa Bradanica Bradanica TroughŽ .fill complex Sabato and Tropeano, 1994 .

The climate of Basilicata ranges from high mountains areas in the western part of theregion with mean annual rainfall in excess of 2000 mm and mean annual temperature of108C to relatively low-lying areas bordering the Ionian Sea with mean annual rainfall of

Ž .less than 700 mm and mean annual temperature in excess of 158C Cataudella, 1987 .The climate of the Salandrella–Cavone area of Basilicata is Mediterranean semi-aridwith more than 65% of rainfall occurring during the autumn and winter months.Summers are characterised by long periods without rain, with temperatures in excess of

Ž408C and relative humidities of less than 40% Servizio Meteorologico Aeronautica.Militare .

Within the main river valleys, agricultural productivity on river terraces is enhancedby irrigation allowing the cultivation of citrus fruit orchards and crops such as tomatoes,

Ž .melons and strawberries Rendell, 1986 . On higher slopes and in the majority of theŽ .river catchments, current land use is dominated by the European Union EU subsidised

Ž .cultivation of durum wheat for pasta , which can be grown on the clay terrain, withouty1 Ž .artificial irrigation, yielding about 2.5 t ha Bianchi et al., 1993 . Most winter cereal

cultivation occurs on slopes shallower than 208 but the use of caterpillar-trackedmachinery has enabled steeper slopes to be cultivated and thus badland features havebeen increasingly confined to marginal scarp slopes. Changing land managementstrategies and increased mechanisation has resulted in the deep ploughing of slopes to

Ž .increase water retention capacity Landi, 1989 . This improvement in mechanisation andan increase in crop specialisation has resulted in economical changes which, in order toimprove the cost–benefit ratio to the farmer, has lead to a general increase in field area,

Ž .removal of terraces and a ‘smoothing of hillslopes’ Zanchi, 1989 . These economicŽincentives have meant that there is a tradition of using bulldozers Calzecchi-Onesti,

. Ž .1954, 1957 and even explosives Pavari, 1911 to remodel and reclaim areas of badlandŽ .landforms in Italy for agricultural productivity. In Tuscany, Guasparri 1993 has shown

that substantial areas of former badlands have been obliterated since the 1960s and inŽBasilicata, major remodelling of badland slopes has occurred since the 1970s Rendell,

. Ž .1986 and is still continuing today see Fig. 7 .Badlands are conventionally considered to be areas of extreme soil erosion. This

paper describes the impact on soil erosion of remodelling these badlands area forincreased agricultural productivity. Land degradation by severe soil erosion has been

Žlinked to desertification in this and other parts of the Mediterranean Basin Thornes and.Brandt, 1996 and it has been suggested that there is an increasing risk of further

Ž .degradation in response to both physical and socioeconomic factors Perez-Trejo, 1994 .The Salandrella–Cavone study area is therefore an ideal location in which to examinethe interaction of these factors in relation to land use change.

( )M.L. Clarke, H.M. RendellrCatena 40 2000 229–250 231

2. Study area

The study area lies between the Basento and Agri rivers, predominantly in theŽ .Salandrella–Cavone river catchment Fig. 1 . The area is underlain by up to 500 m

Ž .ENI-AGIP, 1972 of Plio–Pleistocene clays which are capped in places by locallyvariable, uncemented to partially-cemented, Upper Pleistocene sands and conglomerates.Uplift and differential incision has left pedestal-type outcrops of the conglomerate which

Ž .lie up to 250 m above the river valleys Rendell, 1975 and support hilltop settlementssuch as Pisticci, Ferrandina and Pomarico. At Pisticci, the clays are incised to a depth of200 m relative to the current valley of the Cavone river. Clay mineralogy shows adominance of sodium montmorillonite, although kaolinite and illite are also presentŽ .Rendell, 1975 . The clays are inorganic silty-clays of medium plasticity and disperse

Ž .rapidly when wetted Rendell, 1982, 1986 making them prone to both surface andsubsurface erosion. The grain size of the clays is typically 55–65% by weight finer than

Ž . Ž . Ž .8f 3.9 mm and 90–95% finer than 4f 62.5 mm Rendell, 1975 . Given thedispersive nature of the clays, the high relative relief and the seasonality of the climate,

Žit has been suggested that this area is prone to severe soil erosion Alexander, 1982;.Cataudella, 1987 .

3. Land use change impacts on badland landforms

In order to assess the impact of landscape remodelling on soil erosion it is necessaryto compare the initial and final topographic forms involved in hillslope modification.

Ž .This is extremely difficult because 1 the majority of the landscape remodellingŽ . Ž .occurred prior to 1986 based on analysis of aerial photographs and 2 regional

legislation in 1994 has made remodelling of landforms of geomorphological or geologi-cal interest illegal in order to preserve the landscape heritage of these characteristic

Ž .badland features Legge Regionale, 1994, no. 28 . The imposition of fines is not asufficient deterrent to prevent remodelling particularly when compared with the eco-nomic benefits to the farmer of increasing land productivity, and thus, during 1996–1997we were fortunate to obtain slope geometry data from an area both before and afterremodelling. In all other cases we have chosen to characterise the morphology of

Ž .badland-type sites Serra Pizzuta, Serra del Purgatorio, Mesola della Zazzera and takeŽ .measurements of badland areas adjoining remodelled slopes see Fig. 9 for an example

to compare these data with those for parts of the same slopes that are known to havebeen remodelled based on field observations and analysis of aerial photographs.

4. Characteristics of badland landforms

ŽTwo distinct types of badland landform have been recognised in Italy Alexander,.1982; Sdao et al., 1984; Pinna and Vittorini, 1989 : calanchi are steep gullies with

knife-edge ridges and biancane are small, conical or dome-shaped forms. In order to

()

M.L

.Clarke,H

.M.R

endellrC

atena40

2000229

–250

232

Fig. 1. Map of the study area showing location of sites described in the text.


characterise the typical morphology of these landforms in this area of Basilicata,biancane and calanchi were studied at three sites within the Salandrella–Cavonecatchment. Serra Pizzuta and Serra del Purgatorio lie to the west and south of the hilltoptown of Pisticci, whilst Mesola della Zazzera is located to the northwest of Craco and

Ž .south of Ferrandina Fig. 1 . Measurements of slope length and angles at the sites wereundertaken either by Abney level and tapes or surveyed using a Wild T60 theodolite andstaff.

4.1. Biancane

At the Serra Pizzuta and Serra del Purgatorio sites, biancane forms were surveyed.Slope angles were measured at 0.5-m intervals along north–south and east–westtransects. A total of 26 individual biancane forms were studied, 20 from the SerraPizzuta study area, where biancane forms are developing as a scarp calanchi sloperetreats, and six from Serra del Purgatorio, where the forms are dissociated from scarpretreat. In both areas the biancane are developed on a relatively low angle basalpediment of 5–108. The morphological data for the biancane populations from both sitesŽ .Tables 1 and 2 show a dominance of steep slope angles in the range 30–388. Slope

Ž . Žangles appear not to differ as a function of aspect Table 1 . Mean slope lengths Table.2 varied from 2.7 m to 5.4 m.

Examples of biancane profiles are given in Fig. 2. As the forms get smaller theybecome more symmetrical, with the same steep slope angles maintained on the smaller

Ž .forms. The bases of biancane are marked by a sharp break of slope Fig. 3 and alow-angle pediment covered by redeposited fine-grained sediment. The steepest slopeelements on the biancane, those with slopes in excess of 538, show evidence ofgravitational creep. Biancane surfaces are covered by a typical ‘popcorn’ covering ofweathered bedrock, traversed by a network of desiccation cracks, which form at a

Table 1Biancane slope angles measured in 0.5-m-length slope segments

Slope aspect

N S E W

Serra PizzutaMaximum 638 688 708 628

Minimum 78 18 18 28

Mean"standard deviation 33.4"13.38 35.7"13.18 35.5"14.28 34.3"13.58

Number of slope segments 125 198 184 133

Serra del PurgatorioMaximum 558 688 568 748

Minimum 138 18 48 38


Number of slope segments 46 33 36 45

()

M.L

.Clarke,H

.M.R

endellrC

atena40

2000229

–250

234

Table 2Dimensions and slope lengths of biancane from twenty forms at Serra Pizzuta and six forms at Serra del Purgatorio

Ž . Ž . Ž .Planform m Height m Length of slope m

N–S E–W N S E W N S E W

Serra Pizzuta 7.52"4.60 6.03"3.05 1.72"0.80 2.91"1.79 2.32"1.56 1.82"1.03 3.70"2.57 5.36"3.22 4.21"2.74 3.35"1.54Serra del Purgatorio 5.21"1.56 5.27"1.51 2.32"0.80 1.44"0.83 1.48"0.73 2.24"1.23 3.75"1.33 2.67"0.94 2.88"1.22 3.71"1.55

The mean and standard deviation of measurements from individual biancane are given.


Ž . Ž .Fig. 2. Profiles of two biancane a and b from Serra Pizzuta.

variety of scales. The depth of the weathered layer is typically in the range 10–20 mm,but may reach as much as 40 mm thick, developing in response to seasonal wetting and


Fig. 3. Break in slope at base of a biancane at Serra Pizzuta showing the low angle of the basal pediment.

drying. Biancane surfaces may be covered with small rainsplash pillars which developbeneath marine fossils of bivalves as they weather out of the clay.

Estimates of the rate of surface lowering of the forms were obtained from a series oferosion pins placed along north–south and east west transects on three of the Serra

Ž . y1Pizzuta biancane Table 3 . These show a mean erosion rate of 15–23 mm yr forSerra Pizzuta during 1993–1994 which are similar in magnitude to the 19–27 mm yry1

Table 3Ž .Erosion pin data showing the mean and standard deviation of surface lowering mm found on different aspect

slopes of biancane at Serra Pizzuta in the period between April 1993 and April 1994

Form Number of pins N S E W

Biancana 1 66 19"5 22"4 21"2 16"7Biancana 2 32 23"6 15"9 – 20"6Biancana 3 71 15"4 20"3 20"5 20"5


Table 4Ž .Annual rainfall mm for the period 1980–1988, derived from daily rainfall records at stations in Tuscany and

Basilicata

Tuscany Basilicata

Siena Taverna d’Arbia Monte Oliveto Pomarico Pisticci

Maximum 1048.4 1240.1 974.2 626.9 899.9Minimum 558.2 487.0 495.6 265.10 319.0Mean"standard 785.73"157.06 804.97"220.26 697.26"162.83 413.14"137.40 512.21"181.55deviation

Tuscany data were obtained from the Istituto Idrografico di Pisa and Basilicata data from the InstitutoIdrografico di Catanzaro.

Ž .found at Stazione Craco Fig. 1 in the Salandrella–Cavone catchment during 1978–1980Ž .Alexander, 1982 . The Serra Pizzuta data show no difference in lowering rates as afunction of aspect, or with distance from the biancane summit. This implies thatbiancane erosion at this site is dominated by a process which operates uniformly overthe entire form, i.e., rainsplash. Erosion is detachment-limited which is controlled by thethickness of the weathered layer and rate of movement of the wetting front. Mean ratesof wetting advance of around 7 mm in 90 min have been found for desiccated

Ž .unweathered clays from this area Rendell, 1975 .Many of the biancane show micro-rilling features with dimensions of around 1–5 mm

following topographic low points and randomly traversing bedrock joints, showing thatthere is clearly no lithological control on the pattern of flow networks. These rills mayreflect the collapse of micropiping in the upper 10–15 mm of the weathered crust. Thethin fractured and weathered crust cannot sustain significant piping, although micropipesof dimensions -5 mm may connect crack networks. Field evidence during rainstormevents show that cracks in the weathered surface seal after 7 mm of persistent rain,resulting in overland flow.

Micro-piping has been proposed as the dominant process controlling biancaneŽ .formation in Tuscany Torri et al., 1994; Torri and Bryan, 1997 . This is not the case in

Basilicata. The differences in dominant process may be related to differences in climate.The site studied by Torri et al. in southeastern Tuscany receives between 480 and 1240

Ž .mm of annual rainfall based on 1980–1988 data; see Table 4 , averaging 690 to 805

Table 5Measurements of calanchi slope angles

Calanchi location Serra Pizzuta Mesola della ZazzeraSlope feature Major chutesMajor chutes Minor chutes Interfluves

Minimum angle 218 288 168 198

Maximum angle 548 558 518 65.58


Number of segments 57 32 108 109

Angles measured in 0.5-m-length segments upslope.


Ž . Ž .Fig. 4. Profiles of calanchi at A Serra Pizzuta and B Mesola della Zazzera.

y1 Ž .mm yr in agreement with the values of Torri et al., 1994 and resulting in aŽ .superficial weathered layer thickness of 50–150 mm Torri et al., 1994 , which is at

least five times the depth of the weathered layer found in the more arid area of


Basilicata. Annual rainfall in this region of Basilicata varied from 413 to 512 mmŽ .between 1980 and 1988 the comparable period to the Tuscany data . Depth of the

weathered layer will control piping efficiency, and ultimately affect the overall form ofŽ .the biancane Torri and Bryan, 1997 . Comparison of Basilicata biancane forms with

Ž .those described by Calzolari and Ungaro 1998 in Tuscany appears to confirm thatthere is a genuine difference in form types. For example, the maintenance of form withincreasingly smaller biancane found at Serra Pizzuta and Serra del Purgatorio sites in

Ž .Basilicata contradicts the findings of Torri and Bryan 1997 who argue that in Tuscany,the dominance of micropiping results in a penultimate collapsed souffle-like form.´Biancana form types are defined and described for Tuscany by Calzolari and UngaroŽ .1998 .

4.2. Calanchi

In Basilicata, calanchi forms are seen on erosional terrace scarps of relict fluvialsystems and incised bowl-shaped gully head areas close to hill crests. In this study wehave concentrated on those features associated with the more prevalent erosional scarps.Calanchi were studied at two of these sites on south-facing escarpments at Serra Pizzuta

Žand Mesola della Zazzera. Profiles were obtained in 0.5-m segments Table 5; Figs. 4.and 5 . At both sites gully thalwegs, or chutes, were separated by knife-edge ridges. The

Ž .Serra Pizzuta profiles indicate that chutes and ridges are parallel to each other Fig. 2suggesting that all features on the slope are retreating at the same rate allowing

Ž .maintenance of high slope angles 36.58 to 41.58 . At the Mesola della Zazzera site, the

Fig. 5. The calanchi front at Mesola della Zazzera.


Ž .measured chutes were slightly steeper 438 than those at Serra Pizzuta. Calanchi slopeŽlengths are in the estimated range of 25–60 m being too steep to measure by hand for

.the entire length at overall slope angles greater than 358.ŽIn addition to the geomorphic processes operating on biancane forms wetting and

.drying, rainsplash, micro-piping and micro-rilling , the calanchi in these study areasŽ .show both visual Fig. 6 and morphological evidence of acting as effective debris flow

chutes for weathered material accumulated in the headward regions of their discretedrainage basins. At both study sites the break of slope at the base of the calanchi chutes

Ž .occurred at an accumulation zone forming debris flow cones see Fig. 5 .

5. Slope remodelling

Ž .The remodelling of badland landforms in Italy has involved: i the modification ofŽ .calanchi forms by smoothing out the sharp break of slope at the top head of the

Ž .calanchi and removal of the knife-edge ridges; ii the isolation of calanchi forms,Ž .leaving small areas on otherwise remodelled slopes; iii the obliteration of calanchi

Ž .forms, creating a completely new slope; and iv the obliteration of biancane to producea smoothed slope surface, created by cutting off the tops of the hummocks. With theavailability of increasingly powerful earth-moving equipment, the tendency appears to

Ž . Ž . Ž .have been to favour options ii to iv Calzecchi-Onesti, 1957 . In Basilicata, remod-elling carried out within the last 20 years has involved the implementation of optionsŽ . Ž .iii and iv resulting in the complete obliteration of badland forms.

Ž . Ž .Fig. 6. 3 a Active erosion on calanchi during a rainstorm, Basento Valley southeast of Tricarico. b Evidencefor debris flows after a rainstorm at Serra Pizzuta.


Ž .Fig. 6 continued .

Examples of the kinds of morphological changes involved in remodelling bothbiancane and calanchi in the Salandrella–Cavone river catchment are shown in Table 6.Two examples of remodelling biancane and three of calanchi have been studied and aredescribed below. Observation of these new slopes suggests that they appear to be proneto rilling and gully development.

5.1. Remodelled biancane

At Manca Tortoniero, which lies approximately 2 km to the west of the Pisticci–Cracoroad, two biancane swarms, located on opposite slopes of a drainage divide, were

Ž .observed being actively remodelled in August 1996 Fig. 7a . The site was revisited inŽ .August 1997 Fig. 7b and measurements were taken of the change in slope geometry.

Ž .Both newly remodelled fields one facing ESE and the other WNW showed evidence of

()

M.L

.Clarke,H

.M.R

endellrC

atena40

2000229

–250

242

Table 6The change in slope morphology and characteristic created by hillslope transformations achieved by remodelling in Basilicata

Location Initial badland Slope Initial slope Initial slope Remodelled Remodelled Change Change in slopeŽ . Ž . Ž .landform aspect angles length m slope angles slope length m in angle length m

Manca Tortoniero Biancane swarm ESE 378 -5 168 113 y218 q108Manca Tortoniero Biancane swarm WNW 378 -5 178 99 y208 q94Pantone della Fica Calanchi WSW 408 25 178 90 y238 q65San Bernardino Calanchi W 418 22 128 34 y298 q12La Piana Calanchi SW 468 63 168 90 y308 q27


Ž . Ž .Fig. 7. a Biancane on an ESE facing slope at Manca Tortoniero being actively remodelled, August 1996. bThe same location in August 1997. The abandoned town of Craco is visible on the skyline to the top left of thephotograph.

active soil erosion during the subsequent winter of 1996–1997, exhibiting large rills atŽ .topographic low points in the slope. The rill formed in the ESE facing slope Fig. 8a


Ž .Fig. 8. Large rills formed in the topographic lowpoints of the newly remodelled fields at Manca Tortoniero aŽ .on the ESE slope—note the gully system at the field base b on the WNW slope.

was 101.2 m long and ranged in depth from 0.16 to 0.32 m and width from 0.18 toŽ .0.49 m wide. The rill in the WNW facing slope Fig. 8b was 69.3 m long and ranged in

depth from 0.06 m to 0.28 m and width from 0.09 to 0.32 m wide. Calculations of rill


geometry indicate that erosion of slope sediment and rill formation during the winter of1996–1997 resulted in removal of at least 8.6 m3 and 5.7 m3 of sediment from the ESEand WNW fields, respectively. Inspection of slope surfaces in August 1997 revealed thepresence of networks of pronounced desiccation cracks up to 0.03 m wide and 0.5 mdeep and which may leave the slopes preferentially prone to subsurface erosion bypiping. The large rills on both slopes are interconnected with major gully systems

Žlocated at the field base headcuts currently 2.5 m deep feeding into channels in excess.of 15 m deep which in turn drain into the Salandrella River. Thus, remodelling these

slopes has resulted in the formation of a connecting link with the ephemeral drainagenetwork catchment of the Salandrella–Cavone system. Soil eroded from the large rillsformed in these remodelled fields in winter 1996–1997, can therefore potentially betransported into the perennial fluvial system of the Salandrella–Cavone and out into theIonian Sea. Future monitoring of these fields is planned with particular interest in thepotential of headward retreat of the gully system upslope fed by the connecting largerills and potential pipes fed from large surface cracks. Headward extension of the gullysystems would lead to an irreversible loss in field size, once the erosional forms are toolarge to be ploughed in.

5.2. Remodelled calanchi

Ž .Three remodelled calanchi sites were studied Fig. 1 . All of these sites wereremodelled between 1986 and 1996, based on aerial photograph coverage and fieldsurveillance.

Fig. 9. Remodelled slope at La Piana on the Torrente il Gruso.


5.2.1. Pantone della FicaŽ .The calanchi front at Pantone della Fica faces 2408 WSW with a 408 scarp which is

25 m long. Part of this front has been remodelled resulting in a lower angle slope of 178

and an increase in slope length to 90 m, and has involved the obliteration of a dirt roadwhich ran along the base of the old calanchi front. Formation of a lower slope angle hasresulted in the creation of a new dirt road used by local farm traffic, orthogonal to theold road and running upslope and over the top of the remodelled calanchi, providing acompacted surface for overland flow runoff during storm events. Whilst probably beingthe most direct route, the creation of a road through the calanchi badlands would beimpossible prior to remodelling due to the extreme steepness of the 408 slope. As atother locations in this area where dirt roads traverse slopes, the new road provides anideal focus for future gully development, resulting from the creation of a local drainagesystem and flow convergence on to the road during storm events. Runoff from the roadat breaks of slope or below bends results in sediment entrainment and thus, this site, likemany others on this clay terrain, has a high future soil loss potential.

5.2.2. San BernardinoThe San Bernardino site lies to the northwest of Pantone della Fica and consists of a

remodelled slope adjacent to the remnants of the previously extensive calanchi scarp.Ž .The slopes face 2708 W and are located on the east side of a minor road to Ferrandina.

Slope angle on the calanchi scarp is 418 and the length of slope is 22 m and this has

Ž . Ž .Fig. 10. a A rotational landslide in a remodelled field near Craco after harvest, September 1997. b Shallowdebris flows in a remodelled slope near Grassano, cultivated with durum wheat, April 1996.


Ž .Fig. 10 continued .

been remodelled into a 34-m slope of only 128. The remodelled slope had been used fordurum wheat cultivation during the 1996 and 1997 growing seasons.

5.2.3. La PianaThis site, shown in Fig. 9, lies on the northern bank of the Torrente il Gruso which is

a tributary of the Salandrella River, southeast of Ferrandina. Measurements on theadjacent calanchi scarp to the east of the remodelled slope indicate that bare 63-m slopesof 468 have been converted into 90-m slopes of 168 and cultivated with durum wheat.Wheat growth at this site is very patchy implying low overall yields. Poor landmanagement is indicated by the blue-grey nature of the remodelled field resulting fromthe deep ploughing of bedrock. Nevertheless, an attempt has been made to cultivatewheat on the bedrock without the addition of substantial quantities of organic matter toimprove soil fertility.


6. Slope remodelling impacts on landscape morphology and soil erosion

The slope transformations described above involve increasing slope lengths whilstdecreasing slope angles by more than 208. These changes have repercussions for thespatial organisation and coupling of erosion processes on the slopes. Prior to remod-elling, badland slopes dominated by calanchi or biancane, although visually spectacular,are characterised by very localised processes of erosion and deposition of material. Thebiancane at Serra Pizzuta may exhibit erosion rates of 20 mm yry1 but the materialeroded is deposited adjacent to the erosional landforms. Similarly, the calanchi exhibitvery localised deposition of material carried by mass movement or debris flows. Inmany cases these badlands are either completely or almost completely spatially divorcedfrom local channel systems so that sediment delivery to the channel systems does notoccur. Although the calanchi and biancane forms are mostly unvegetated, they aresurrounded by areas of scrub vegetation. After remodelling, ploughed slopes remainunvegetated, this is critical during the late summer and early autumn periods and whenthey are particularly vulnerable to the type of convective rainstorms that characterise theend of summer in central and southern Italy.

The slope transformations increase the effective catchment for rainfall by reducingŽslope angle a change from 408 to 158 for the same slope length results in an increase of

.26% in catchment area for rain falling vertically and increasing slope length. Whereasdetachment of material by rainsplash is the dominant process on the biancane, on theremodelled slopes rainsplash, ponding and overland flow create process interactionsdelivering sediment off-slope and potentially off-site. As has been shown, the new slopelengths will facilitate the development of concentrated overland flow with the develop-ment of rills and gullies delivering water and sediment downslope and off-field intowater courses and on to roads.

Another impact of remodelling is an apparent increase in the occurrence of relativelyŽ .small-scale rotational landslides Fig. 10 and shallow debris flows, with typical depths

of 0.8 m. These tend to be sufficiently superficial in morphology that, like many rills,they can be ploughed in at the beginning of each cultivation cycle and are thereforerecurrent but short-lived landscape features. Deep ploughing techniques used in this areaturn over sediment to a depth of 1.4 m.

7. Implications of changing land use policy

The practice of remodelling badland hillslopes to increase agricultural productivityfrom the clay landscape has been exacerbated by the provision of EU subsidies tocultivate durum wheat. The pattern of subsidy changed in June 1992 with the introduc-tion of a staged reduction in intervention price for durum wheat. This change has beencoupled with a dual subsidy system whereby for 1995–1996, a hectarage payment ofaround 643,000 lire hay1 was given with an additional yield-related value of 163,000

y1 Ž y1 .lire ha for a yield of 3 t ha which amounts to at least 40% of farm incomeŽ .Bianchi et al., 1993 . The continuing hectarage-based economic incentive to increaselandscape productivity has inadvertently resulted in an increase in soil erosion in these


areas, where comparatively stable badland landforms have been replaced by lower angleagricultural fields which have a higher potential for soil erosion by off-site removal ofsediment. The issue of cereal subsidies has, to some extent, been balanced by theprovision of subsidies to farmers to maintain areas of fallow land by the introduction of‘set-aside’. The decision by the EU in 1997 to discontinue subsidising set-aside willhave a negative effect on measures to prevent soil erosion, as it becomes economicallymore beneficial to cultivate all fallow land. Economic incentives may result in a furtherincrease in the number of illegally remodelled badland slopes. The use of subsidies forspecific crops may also increase the likelihood of soil erosion. For example, in the 1997growing season, there has been an increase in the number of fields in the study areagiven over to the cultivation of sunflowers. The percentage ground cover of crops likesunflowers is low and thus, the potential for soil erosion during a growing season isgreater than for a larger ground cover crop such as durum wheat. Land use policy and itssubsequent economic implications are thus important in determining some of thecontrolling factors on soil erosion.

8. Conclusions

In Basilicata, the practice of using heavy earth-moving machinery to remodelcalanchi and biancane badland landforms changes slope morphology. Initially steepbadland scarps of )358 are lowered by 20–308 to form more gentle slopes of 12–178

which can be ploughed and then cultivated for cereals. Slope lengths are also greatlyincreased resulting in a spatial coupling of soil erosion processes and an extension ofexisting drainage networks. This change in land use, promoted by economic incentives,has resulted in an increase in potential soil erosion in this area.

The effects of ‘soil quarrying’ for infrastructure improvement or land preparationŽ .may be considered to be soil degradation processes Poesen and Hooke, 1997 in that

soil is physically removed from the landscape or locally repositioned on it. However, itis the geomorphological and associated hydrological changes to the slopes, as a result ofthese landscape remodelling activities, that have the potential for significantly increasingsoil erosion and land degradation.

Acknowledgements

We would like to thank Dott. F. Belelli of the Unione Provincale Agricoltori di Sienaand Dott. C. de Giovanni of the Istituto di Allevamento delle Piante, Bari for technicaldata, Geom. G. Alfieri of the Istituto Idrografico di Catanzaro for meteorological datafrom Basilicata, and the Santangelo family of Tricarico for their continuing hospitality.

References

Alexander, D.E., 1982. Difference between ‘calanchi’ and ‘biancane’ badlands in Italy. In: Bryan, R., Yair, A.Ž .Eds. , Badland Geomorphology and Piping. Geo Books, Norwich, UK, pp. 71–87.


Bianchi, A., Boggini, G., Di Fonzo, N., Pasquini, M., Pogna, N.E., Ronga, G., Azzoni, A., 1993. La rispostadella durogranicoltura italiana alla nuova politica agricola comunitaria. L’Informatore Agrario 34, 27–33.

Calzecchi-Onesti, A., 1954. Sistemazioni del Terreno e Fertilita in Collina. Ramo Editoriale Degli Agricoltori,`Rome, 107 pp.

Calzecchi-Onesti, A., 1957. Sistemazioni in Collina. Ramo Editoriale Degli Agricoltori, Rome, 350 pp.Calzolari, C., Ungaro, F., 1998. Characterization and quantitative spatial analysis of geomorphic features of a

Ž . Ž .badland biancane area central Italy . Catena 31, 237–256.Ž .Cataudella, M. Ed. , 1987. Altante della Basilicata: il Territorio per Immagini. Regione Basilicata, 47 pp.

ENI-AGIP, 1972. Acque Dolce Sotteranee. Milano.Guasparri, G., 1993. I lineamenti geomorfologici dei terreni argillosi pliocenici. In: La Storia Naturale della

Toscana Meridionale. Monte dei Paschi, Siena, pp. 89–106.Landi, R., 1989. Revision of land management systems in Italian hilly areas. In: Schwertmann, U., Rickson,

Ž .R.J., Auerswald Eds. , Soil Erosion Protection Measures in Europe. Catena Soil Technology Series, 1, pp.175–188.

Legge Regionale, 1994. Individuazione, classificazione, istituzione, tutela e gestione delle aree naturali protettein Basilicata. 28 June 1994, No. 28.

Pavari, A., 1911. Le crete senesi e lo scasso con la dinamite. L’Italia Agricola, Piacenza.Perez-Trejo, F., 1994. Desertification and land degradation in the European Mediterranean. EUR 14850 EN,

Office of the Official Publications of the European Communities, Luxembourg.Ž .Pinna, S., Vittorini, S., 1989. Su alcune caratteristiche delle argille plioceniche della Valle Dell’Era Toscana

in rapporto alla genesi di calanchi e biancane. Geografia Fisica Dinamica Quaternaria 12, 131–137.Poesen, J.W.A., Hooke, J.M., 1997. Erosion, flooding and channel management in Mediterranean environ-

ments of southern Europe. Progress in Physical Geography 21, 157–199.Radina, B., 1964. Contributo alla conoscenza del dissesto geologico del versante Ionico Lucano. Bolletino

Societa Naturalistica, Napoli 73, 211–265.Rendell, H.M., 1975. Clay hillslope erosion in a semi-arid environment: the Basento Valley, southern Italy.

Unpublished PhD thesis, University College London.Rendell, H.M., 1982. Clay hillslope erosion rates in the Basento Valley, S. Italy. Geografiska Annaler 64A,

141–147.Rendell, H.M., 1986. Soil erosion and land degradation in southern Italy. In: Fantechi, R., Margaris, N.S.

Ž .Eds. , Desertification in Europe. Commission of European Communities, Brussels, pp. 184–193.Sabato, L., Tropeano, M., 1994. Cenni sul sistema catena–avanfossa–avampaese in Italia Meridionale. In:

Guida alle escurzioni. Congresso. Soc. Geol. Italia, Bari, 1994. Quaderni Bibliografico della Provincia diMatera, 15, pp. 11–32.

Sdao, G., Simone, A., Vittorini, S., 1984. Osservazioni geomorfologiche su calanchi e biancane in Calabria.Geografia Fisica Dinamica Quaternaria 7, 10–16.

Thornes, J., Brandt, J., 1996. Mediterranean Desertification and Land Use. Wiley, Chichester.Torri, D., Bryan, R., 1997. Micropiping processes and biancane evolution in southeast Tuscany, Italy.

Geomorphology 20, 219–235.Torri, D., Colica, A., Rockwell, D., 1994. Preliminary study of the erosion mechanisms in a biancana badland

Ž .Tuscany, Italy . Catena 23, 281–294.Zanchi, C., 1989. Drainage as a soil conservation and soil stabilizing practice on hilly slopes. In: Schwert-

Ž .mann, U., Rickson, R.J., Auerswald Eds. , Soil Erosion Protection Measures in Europe. Catena SoilTechnology Series, 1, pp. 73–82.

The impact of the farming practice of remodelling hillslope topography on badland morphology and soil erosion processes

Documents