Annals of Arid Zone 35(3) : 225-240, 1996

Sand Dune Stabilization in the Thar Desert of India: A Synthesis

R.N. KaulF-B, SOli tit Extension, Pan-II, New Dellti 110 049, India

Abstract: Sand dune stabilization programme in the lbar desert of India is basedon an understanding of the edaphic characteristics of the dune-interdune systems,as well as on the soil-water-plant relationships of the major dune types. A numberof physical and chemical methods of stabilization have been tried during the lastfour decades. It has been found that the more lasting effects are achieved througha careful plantation of trees, shrubs and grasses at appropriate sites and usingproven technologies. lbere is, however, a need to further refine the techniquesto suit the different dune environments and to make the venture economically moreviable.

Key words: Thar desert, dune ecology, chemical stabilization, physical stabilization,biological stabilization.

Sand dunes cover extensive area of theThar desert which lies between the Aravallihill ranges in the east (in Rajasthan stateof India) and the Nara river in the west (inPakistan). In India, the northern limit ofthe Thar is along the Indo-Gangetic plainsin the states of Haryana and Punjab and itssouthern limit is along the north Gujarat plains.Sand dunes of many different types have beenrecognised in the Thar, the major ones beingthe longitudinal (linear), transverse, parabolic,network, star, barchan and barchanoid, withheight ranging from - 2 m to 50 m or more(Kar, 1993). Low sand streaks and sandyhummocks are also numerous. Based on theirage of formation, sand dunes have been clas-sified as the dunes of 'old system' and 'newsystem' (Pandey et al., 1964). UsualIy thehigh sand dunes are of old system. Thesewere formed in an earlier dry climate andhave a greater stability than the dunes ofnew system which are forming now.

The major causes of the formation of manylow sand streaks, sandy hummocks and bar-chans are over-grazing, faulty agricultural

practices, including ploughing along the duneslopes, as welIas destruction of natural vegeta-tion cover on dunes and sandy plains forfuel and fodder. Such activities are also lead-ing to the advancement of old dunes in manyparts of the desert. In fact, very high humanand livestockdensities in the desert are leadingto mismanagement of the sandy terrain, caus-ing sand reactivation and land degradation.The total area affected by sand drift in Rajas-than, Gujarat and Haryana states is estimatedas 88,078 sq. km. The ultimate result is sandencroachment upon productive agriculturalfields, human habitations, canals, roads andrailway tracks. In order to control the menaceof sand drift it is necessary to undertake ef-fective sand dune fixation programme.

Ecology of Sand Dune Ecosystem

For successful fixation of inland sand dunesit is necessary to understand the ecology ofsand dune system,particularly factors affectingsoil-water-plant relationship. Success ofbiological sand dune fixation largely dependsupon the delicate balance between theavailability of soil moisture and its use forplant growth over a period of time. This

226 KAUL

balance has to be considered in the contextof climatic, edaphic and hydrological factorsvis-a-vis plant adaptability as it influences thechoice of plant species and the planting tech-niques, viz., direct seeding vs. transplanting,age of seedling, spacing, depth of planting,etc.

Climatic characteristics

The climate in the Thar desert is char-acterized by extremes of variation in diurnaland annual temperatures, scarce and erraticrainfall and high evaporation. Drought, dueto lack of moisture, is quite frequent. Mon-soon arrives here by the last week of Juneor first week of July. The annual rainfallvaries from as low as 100 mm to 150 mmwith 11 dry months in the west (Jaisalmerin Rajasthan) to 400-450mm with 9 dry monthsin the east (Sikar in Rajasthan; Mahendragarhin Haryana). Bulk of the rainfall occurs duringthe south-west monsoon season between Juneand September with a peak in August. Al-though not much rainfall is received in thedesert, the atmospheric humidity is unusuallyhigh and is comparable to that in places ofhigher rainfall in the semi-arid and sub-humidzones (Krishnan, 1%8). The mean monthlytemperature during the hottest months (Mayand June) is 40°C, but it often goes as highas 45°C which is very detrimental for plantgrowth. In this period dry and hot dust-raising\\~nds,popularly known as '100' and duststorms(andhi) 'occur very frequently (Krishnan,1977). The potential evapo-transpirationduring summer varies from 7 to 9 mm day-I.Winter is very cold. Normal minimumtemperature varies from 1°C to sac and inexceptional cases, it goes down to as lov.'as -4°C. In severe winter, frost occurs fre-quently.

Edaphic characteristics

Physical: Dune soils are single-grain, non-coherent and structureless, and are, there-

fore, highly erodible. Their apparent specificgravity ranges from 1.62 to 1.75 as comparedto 1.4 to 1.5 in the case of other soils. Thetotal pore space is, therefore, less in the dunesoil and the pores are of very large size.

As a result, during rainfall or irrigation,water movement is free and rapid (infiltrationrates being 10 to 16 cm h-1), leading to heavypercolation losses. Water retention capacityof dune sand is very low, ranging from 4.0to 6.0% at 0.1 bar and 1.5% at 15 bar. Theavailable water holding capacity of the dunesand ranges from 3.0 to 4.5% (w/w).

Pores being of large size in sand, thereis negligible unsaturated hydraulic conduc-tivity; these are too big to hold any wateragainst the force of gravity. Therefore, theonly places where water can be held in sandare the spaces where pores acquire sufficientlynarrow size. This is possible at points ofclose contact' of sand grains with each other.Since such pore spaces account for only asmall fraction of the total pore space, thedune soils have field capacity values of 3.5to 6.0% only. This gives them a moistureretention capacity of 50 to 90 mm m-1 depth.Of this, 12 to 20 mm is too tightly held tobe available to plants. The maximum waterthat can be retained by these soils is 38 to70 mm m-l depth (Dhir, 1985).

There is hardly any crust formation onshifting sand dunes due to low percentageof silt and clay, whereas, on stabilized andsemi- stabilized sand dunes, soil crust isformedbecause of the presence of organic matterand algal growth. Gupta (1976) reportedthat the soil crust so formed on these sanddunes at Bikaner attained a strength of 3to 4 kg cm-2 which provides stability to thesedunes.

Chemical and microbiological: Eventhough insufficiency of soil moisture is a limit-ing factor in plant establishment and subse-quent growth, soil fertility is important for

SAND DUNE SfABILIZATION

Table 1. Cenaill chCrnical characteristics of dUlle soils of Bikaller

227

Locality/stabilized& unstabi-Iized dune

Beechwal

Stabilized

0-30 em

30-60 em

Unstabili:z:ed

0-30 em

30-60 em

Sheobarl

Stabilized

0-30 em

30-60 em

Unstabilized

0-30

30-60

Shrlkolayat

Stabilized

0-30 em

30-60 em

pH

8.7

85

8.9

85

8.6

8.8

8.7

8.8

8.7

8.7

E.C.(mmhos

em'l)

0.07

0.07

0.07

0.08

0.09

0.07

0.08

0.07

0.09

0.09

O.c.(%)

0.10

0.04

0.30

0.03

0.12

0.04

0.03

0.02

0.09

0.07

N(%)

0.009

0.014

0.006

0.011

0.015

0.004

0.006

0.006

0.014

0.D15

C/N(ratio)

9.9

3.6

5.0

2.7

7.8

9.8

5.2

2.7

6.44.7

P(%)

0.013

0.020

0.013

0.020

0.020

0.025

0.025

0.007

0.013

0.025

NH4+ N03N (ppm)

30.8

22.4

'30.8

26.0

35.2

22.4

28.0

28.0

30.8

25.2

NH4+N03 Nof totalN(%)

34.0

16.0

50.0

16.8

51.0

46.6

46.6

22.0

16.8

Unstabilized

0-30 em 8.2 0.10 0.02

30-60 em 85 0.12 0.03

Source : Aggarwal and Lahiri, 1981.

0.005

0.007

4.6

4.40.025

0.020

33.6

36.4

67.0

52.0

an optimum utilization of this limited moisture.Aggarwal and Lahiri (1981) reported that inboth the stabilized and unstabilized dunesEC was comparable, but pH varied between8.2 and 8.9 (Table 1). Although the organiccarbon content was generally low in stabilizeddunes, it was consistently more than that inunstabilized dunes. The surface soils of sta-bilized dunes had marginally higher organicmatter compared to that in the subsurface.The C/N ratio was nearly 10:1 in the surfacesoils of stabilized dunes, but quite low inthe subsurface layers (2.7 to 4.7). This suggestsslow, but a definite trend of soil fertility im-provement, at least in the surface layers of

the stabilized dunes with a possible in-volvement of microbial activities, despite theprevalence of high temperature, low soil mois-ture (Lahiri, 1964) and hazards of winderosion. Venkateswarlu and Rao (1981)showed an increase in C/N ratio in stabilizeddunes. Population of fungi, actinomycetesand bacteria were considerably higher in thestabilized dunes than in the unstabilizcddunes.

A study by Dhir and Gajbhiya (1973) onthe distribution of various major nutrient ele-ments revealed that the soil in the immediatevicinity of grass clumps had nearly 25 to 40%

---.----- _________ . H_ .. _ .. _,

228 KAUL

Table 2. Nutrients dcpleriollladdirioll as affecrcd by salld mOl'cmelll under diffcrelll land use colldiriolls from 5..f.7710 2-1.6.77

Localityl Sand depletion/ Nutrients (kg ha'l)Treatment addition (t ha'l) Organic Total Mineral Total Total

matter nitrogen nitrogen phosphorus potassium

Beechwal

Dare sandy plain -1449.0 -724.5 -115.90 -44.9 -231.8 -2159.0

Pearl millet stubble -22.5 -38.3 -2.25 -0.7 -3.8 -34.9

Grass + 13.5 . +35.8 + 1.62 +0.4 +2.7 +22.4

Udalramsar

Dare unstabilized -5560.5 -2780.0 -333.60 -166.8 -1000.0 -6282.8sand dune

Stabilized sand dune + 151.5 + 2121.0 +227.30 +30.3 + 333.3 + 1818.0

more humus, phosphorus and potash thanin the s'oil away from the rhizome. Underan arid environment, the plants, particularlythe trees, utilize nutrients perhaps from thedeeper soil layers, and enrich the fertilityof surface soil, possibly through a nutrientcycling mediated by vegetal residues. In otherwords, the soil in the immediate vicinity ofthe plant has higher contents of available formof various nutrients. Studies by Gupta andAggarwal (1980) showed that maximum sanddepletion took place from bare unstabilizedsand dunes (5560.5 t ha-l), followed by baresandy plains (1449.0 t ha-1). Fields with grasscover and stabilized sand dunes recorded ac-cumulation of sand (13.5 t and 151.5 t ha-l,respectively). The trends were similar in thecase of organic matter, total N, mineral N,total P and total K (Table 2).

Soil-water-plant relationship of shifting alld sta-bilized dUlles

Singh and Shankarnarayan (1986) reportedhigher moisture on the leeward side of sanddunes as compared to their windward side.Moisture content was also higher at the crest.These variations may be due to the natureand thickness of sand and the size of sandgrains. Sand dune fixation studies carried

out in the late 50s and the early 60s (Bhimayaet al., 1961; Kaul, 1970) revealed that soilmoisture regime in shifting dunes (withoutvegetation) was more favourable as judgedby the increased establishment and growthof the planted seedlings, compared to thesemi-stabilized (with vegetation) dunes. Theaccumulation of soil moisture in shifting dunesmay be due to absence of transpiration lossesand the mulching action of the loose surfacesand (Lahiri, 1977). Studies by Krishnan etal. (1966) revealed that in the unstabilized(shifting) sand dunes, a sharp discontinuityin moisture content occurred in the layer 1.5to 2.0 mm below the soil surface. Belowthis layer, the soil moisture reached morethan 5% by weight (i.e., field capacity)throughout the year. However, in the sta-bilized sand dunes, the moisture content upto 3.0 m depth was generally less than 1.5%,although the moisture content, as such, in-creased with depth. The moisture below 3.0m profile in stabilized dune was highly variabledue to the differential extraction pattern oftrees, shrubs and grasses over the stabilizedsand dunes.

Since the moisture status of the profileis influenced by the rainfall received duringthe rainy season (i.e., July to September),

SAND DUNE SrABILIZATION 229

soil moisture contents during the rainlessperiod (i.e., October to June) are of particularinterest. In unstabilized (shifting) dunes,reasonably high soil moisture may be en-countered within 5 to 30 cm of the surface,while in stabilized dunes, moisture contentabove the permanent wilting point may befound (Mann el 01., 1976; Table 3). Thus,unstabilized dunes offer better soil moisturecondition for plant establishment and growthand also support the usefulness of deep plant-ing in shifting sand dunes as is being practicedin Israel (Kaplan et 01., 1970) and in Egyptand Yemen (Costin el 01., 1974). Deep plant-ing also provides favourable temperature forroot growth during summer at the early stagesof plantation.

The low unsaturated state hydraulic con-ductivity of sandy soil not only influences the

rate of evaporation but also the rate of soilmoisture movement within the profile andits utilization by vegetation. Since the inter-site movement of moisture is very small,· thedegree of moisture utilization in dunes willbe, to a large extent, dependent upon thedegree of direct access of the roots to thepoints of entrapped moisture. Bhimaya andKaul (1965), while studying root systems ofAcacia senegal, Albizzia lebbek, Prosopiscineraria and Tecomella IIndlllala, reportedthat the rooting pattern of these species indifferent soil types was somewhat different.It was also observed that the root morphologymay change with changes in environmentalconditions. On undisturbed dunes, Kaul(1970) observed that Acacia lortiUs, whichdevelops lateral roots at a much greater depthin other soils, developed a deep root system

Table 3. Soil moisture (%) slatus of unslabi/ized alld slabilized dUlIcs al differelll limes of Ihe )'ear ill Banner(recorded 1963-6/)

Depth March September June January(cm) Unsta- Stabilized Unsta- Stabilized Unsta- Stabilized Unsta- Stabilized

bilized bilized bilized bilized

0-5 0.3 OJ 0.0 0.1 0.4 0.8 0.9 0.1

5-15 0.9 0.3 1.9 0.5 0.8 0.5 0.8 0.3

15-30 3.2 0.8 2.3 0.3 1.5 0.8 1.5 0.1

30-45 2.8 0.4 2.8 0.7 1.7 0.9 2.4 0.4

45-60 2.6 0.8 3.1 0.8 2.1 0.8 2.1 0.4

60-75 3.5 0.5 2.6 0.6 2.7 1.4 2.4 0.3

75-90 2.8 0.7 3.3 0.9 3.9 1.7 2.2 0.6

90-105 2.6 1.1 4.3 0.9 2.2 1.3 3.4 0.4105-120 3.3 1.6 3.8 1.1 4.3 0.9 4.2 0.8

120-135 3.2 1.7 4.1 1.3 3.3 1.0 4.8 0.7135-150 3.6 1.5 4.2 1.6 3.4 2.6 3.7 0.9150-165 4.1 2.0 5.1 1.5 3.2 3.1 4.2 1.2165-180 4.7 1.8 5.8 1.4 3.5 2.0 5.5 1.3180-195 5.1 2.1 5.0 1.6 4.1 2.2 4.2 2.2195-210 4.8 1.9 5.1 1.7 5.0 1.9 5.0 2.3Moisture content (mm) 41.0 10.2 44.8 10.3 33.3 16.9 33.5 6.6upto 105. cm depth

Moisture content (mm) 105.6 38.5 119.5 33.3 93.6 47.9 104.9 27.7upto 210 cm depth

Source : Mann 1:1al., 1976.

230 KAUL

with extensive spread oflateral roots relativelyclose to the surface, perhaps to make useof surface moisture for initial establishmentand active growth. During the remainingdry months the deep root system helps inextracting moisture from deeper layers.Therefore, plants having a fibrous root systemare better suited to sandy habitats.

Vegetation

Various stages of development or degrada-tion of plant communities on sand dunes areencountered in the Thar desert. Out of thosestages, Saxena (1977) reconstructed the suc-cessional pattern of natural vegetation (Fig.1). According to him, Cypenls arenarius, Aris-tida fil1liculata, Cenchnls biflonls, Tribulus ter-restris, Citrullus colocynth is, Indigoferacordi folia, I. argentea, Farsetia hamilton;;,Crotalaria burlJia, and Aerva pseudotomentosaare the colonising species. Initially, a large-scale coverage by Crotalaria, Aerva andCypenls sps. brings about the stabilization ofsand to a great extent. This makes the sub-stratum more suitable for succession of un-dershrubs, shrubs and perennial grasses, e.g.,Sericostemma pauciflorum, Leptadeniapyrotechnica, Clerodendron phlomoides, Cal-ligonltm polygonoides, Calotropis procera,PaniCllm tllrgjdllm, P. antidotale, LasizlnlS sin-dicllS and CenclmlS ciliaris. Subsequentstabilization a9d undisturbed conditions bringabout Acacia jacqltemontii, Lycillm barbanlm,Balanites aegyptiaca and May tenus emar-ginatlts. The last three species form the penul-timate stage for the climax community ofProsopis cineraria. The grassland develop-ment in the low-rainfall zOl1e(below 300 mm)gets arrested up to Panicllm tllrgjdllm typeonly, whereas, in the higher-rainfall zone(above 350 mm), the same stage is surpassedbySacchanl11l species. The highest grass com-munity is represented by Sacchanl11l ben-golense.

From the foregoing discussion, it is ap-parent that with increasing depth in the unsta-bilzed dunes, soil nutrient depletion is greater,soil moisture content is higher and soiltemperature decreases faster, as comparedto those in the stabilized dunes. Likewise,soil fertility also declines more with depthin the unstabilized dunes. These observationsare of considerable significance for furtherrefinement ofthe technology for greening sanddunes.

Stabilization of Sand Dunes

Prerequisites for sand accumulation anddune building include wind velocity abovea threshold that induces sand movement,availability of sand prone to erosion andpresence of an obstruction around whichmoving sand tends to accumulate. The for-mation of dunes may be prevented or sloweddown by adopting measures like: (a) reduc-ing the velocity of wind by erecting barriers,(b) restricting sand movement through chemi-cal sprays, and (c) encouraging the develop-ment of vegetation that binds sand.Depending on the convenience and theavailability of materials, dunes have been sta-bilized through use of water, chemicals andsealants, erection of physical barriers andthrough afforestation.

Water is a fairly good stabilizer for shortperiod. Frequent watering leads to formationof crust that can tolerate erosive force ofaverage wind speed. This practice may beadopted in regions where water is availableand the area to be stabilized is limited.

Chemical stabilization

Sand dunes have been fixed through spray-ing of byproducts of petroleum on dune sur-face, use of chemical mulches and sealants.Ben Salem (1985) reviewed the techniquesand the products available for sand dunestabilization.

SAND DUNE STABILIZATION 231

/ ~~:'Maytenus emarginatusBalanites aegyptiacaAcacia jacquemontii

!tLycium barbarum

~tLeptadenia pyrotechnicaCalotropis procera

11

Grassland(200-350 mm)

1ICymbopogon spp.Lasiurus sindicusCenchrus sp.

itCyperus arenarius

HLoose sand

EleusineDactylocteniumindigofera

/1/

HAristidil spp.Cenchrus spp.

Panicum turgidum

Saccharum bengalenseS. spontaneumPanicum antidotale

11

Eragrostis spp.Cyperus arenarius

+tBare sand dune

~tWoodland(200-350 mm)

~tClerodendroll phlomoidesCalligonum polygonoidesPanicum turgidum

Cenchrus bij10rusAristidil funiculataIndigofera linii

Serixostemma paucij10rumCrotalaria burhiaAerva pseudotomentosa

j I!l

Hala'C)'lon salicomicumPanicum turigidum

Calotropis proceraCalligonum polygollOides

Dipterygium glaucumSericostemma paucij10rumAerva pseudotomentosa

Acacia jacquemontii

Fig. 1. Stages of developmentldegradiltion of plant communities on sand dunes(Source: Saxena, 1977).

232 KAUL

Byproducts of oil illdlistry : In regions whereoil is available, byproducts of oil industry havebeen used to stabilize the dunes. Productsused include: (i) low gravity asphaltic oil(heavy oil), (ii) high gravity (light) waxy oil,and (iii) crude oil. The asphaltic oil formsa thin, non-sticky sheet over the dune surface,but is easily damaged by animals. Repeatedspray is often required. It is usually usednear the roads where tree plantation is notpossible. High gravity waxy oils are relativelymore effective and long lasting. Such oilspenetrate the 10to 20 cm surface layer. Crudeoil is often sprayed to stabilize the dune sur-face. It may be effective for 2 to 3 yearsby which time vegetation may take up theprimary role of dune stabilization. In 1963,an oil company (ESSO Ltd.) carried out ademonstration at Udairamsar near Bikanerand at Jaipur. After protection, the duneswere planted with grass and tree saplingsand oil named SDSO was sprayed on thedune surface. The crust formed due to SDSOprevented soil movement, but was prone todamage due to trampling by grazing animals.The cost of such treatment in 1963 was Rs.184ha-1 as against Rs.197 ha-1 for conventionalvegetative method developed by CAZRI

(Bhimaya et al., 1961). A number of otherchemicals are available which can be usedto stabilize the sand surface (Table 4).

Other chemicals : Some chemicals aredesigned to facilitate growth of vegetationon unstable surfaces. The time for disintegra-tion of these chemicals depends on the soilstructure, slope, spraying techniques and con-centration of the chemical. Most of thesehave been applied in Europe and Americafor -stabilization of slopes in road cuts orfor fIxation of coastal dunes with vegetation.

These chemical products are available inthe market and may be employed to stabilizethe surface for short time and till much moreeffective vegetal cover has developed. In Mas-tung valley in Pakistan, sand dunes have beenstabilized by spreading soil binding materialslike fIne clay, mud and water mixed withcarbonate-rich fIne clay (Mohammad, 1988).

Physical stabilizatioll

Locally available materials have been usedto erect physical barriers to minimize thewind velocity and prevent sand movement.These usually include parts of locally availableplants or other wastes like rail road sleepers,

Table 4. Major chemical products al'ai/able in the market and their characteristics

Uresol 156 E Colourless liquid

Market FormnameSand stop Powder

Application

Spray

Spray

Rate ofapplication20-25 g m·2

Concentration inwater between 3.4and 6.4%; 1.5L m'2

Effect

Powder absorbs moistureand forms a coatingPenetration from 4-5 mmaggregate formation

Uresol 310 EA Clear, pale yellowHulus 801 Dark-brown concentrate

of white emulsion

Agro fIX

Unosol

Texand D

Milk)' whiteLatex base product

Mesh of very fine strands

SpraySpray 10-40 g of

concentrate for 2 Lof water 10-15 gm·2

Spray 30-50 g m·2

Spray mixed with 150 g m·2crude oil and afterdilution with water

Solidifies within 2 hours byreacting with S02

SAND DUNE SfABILIZATION

Table 5. Pattern adopted and materials used to erect microwind breaks in certain countries

233

Country DesignIndia Checkerboard

(2 to 3 m2)

Iraq Checkerboard(2 to 5 m2)

Libya Checkerboard

Madagascar Parallel strips(Coastal) 250 cm apartPoland CheckerboardRomania Parallel rowsSomalia Parallel rows

(15 m apart)Sudan Checkerboard

(6 sq. m)SyriaTunisia Parallel ro\'/S

(200-300 m apart)Yemen Parallel rows

(20 40 m apart orcheckerboard)

Burrying 15 cm deep, stalks of dry grass/stubble. Imperata cy/indrica,Aristida pungellS, Artemisia herba-alba0.75 to 5.09 m high Aleos sisa

I m high palisades of local brushwood, straw, etc.Reed fencesLocal plant material, Commiphora bilderbrandtii, C. muyrrha, C. gowlelloand Besamotnamnus busseanusDry branches of Leptadenia pyrotechnica

Branches of Pinus brautia used for mulching dune surfacePalm branches, now often replaced by asbestos, cement corrugated sheets

Dry reed fencing (palisades), 0.5 m high

telephone poles, used oil drums, etc. Whenthese materials ale inserted vertically in theground, called 'palisades', they act as micro-wind breaks and when spread over the surface,called 'thatching', they act as mulching andprevent sanj movement. Kaul (1985a) listedthe different types of barriers being used theworld over (Table 5).

Mechanical dune ftxation is effective andhas been applied over extensive areas in mostof the developing countries. It is, however,labour-intensive and is therefore slow. Incountries where labour is in short supply,mechanical dune ftxation will prove costly.

Biological stabilization

Establishment of vegetative cover on sanddunes is often thc best, most permanent andthe most effective method of stabilization.Before any attempt is made for afforestation,it is imperative to undertake preplantingmeasures so that sand movement is minimized.

These measures prevent burial of seedlingsdue to sand deposition and exposure of rootsdue to soil loss.

Planting technique : Before planting ofvegetation on sand dunes, the usual practiceis to erect long parallel barriers (or micro-windbreak) of low height, using locally avail-able plant material. The barriers are putat 5 m to 10 m interval across the prevailingwind direction. Where wind direction is vari-able, cross-barriers are also erected, thuscreating a grid pattern. The pattern, i.e.,distance betwecn parallel lines and the sizeof checker board and the height of barriers,depends on a number of factors like velocityof wind, steepness of slope and type of sanddune. The sand within the grids/squares isusually stable enough to allow establishmentof the transplanted seedlings. At times castoris sown as a nurse crop. It serves as a micro-windbreak, and in addition provides extra in-come through sale of castor seeds.

234 KAUL

In the interdune plains sand movementis not much and, therefore, no micro-windbreak is erected.

Species selection : While selecting plantsfor afforestation the species having the fol-lowing characters are desirable: (a) a mixtureof plants having deep vertical roots to tapmoisture from lower moist zone, and thosehaving surface roots that can take advantageof moisture in the surface layers after lightshowers, and at the same time, possess highroot binding index, (b) ability to withstandabrasive action of blown sand and high windvelocity without being uprooted, (c) abilityto tolerate extremes of temperature, both frostand hyperthermia, and (d) capability of selfregeneration. Kaul (1985b) listed trees,shrubs and grasses commonly used in affores-tation programmes in different countries.

Seedling production : In deserts, the periodof favourable soil moisture regime is ex-tremely short. To take maximum benefit ofthis favourable soil moisture, it is desirableto develop a nursery to raise the seedlings.It should be ensured that regular water supplyis available in the nursery and that it is nearthe site of plantation. Seedlings may bedeveloped, depending on their availability, inpolythene tubes or earthen tubes or in beds.The sowing in nursery should be done soas to get saplings of desirable age/height atthe time of transplantation in July.

Planting time: In order to ensure a highersurvival rate, the planting of seedlings inHaryana and in Indira Gandhi Nahar Pariyojna(IGNP) Stage II areas in Rajasthan is doneafter the onset of rainy season (July to Sep-tember), when the sand is moist. In IGNPStage II areas, planting is done from Julyto October. It can even be done as earlyas April and May, since adequate irrigationfacilities are available, and take the full advan-tage of' the following rains.

Planting depth: Experiments carried outat Bikaner have shown that planting 35 to40 cm deep resulted in higher seedling survival(Kaul, 1983). In Haryana and Rajasthan,the standard practice is to plant in pits 40-45cm deep. Since soil moisture increases withdepth, it is important that planting be doneas deep as possible, depending upon thetolerance of species to deep planting. Forexample, in Israel, large-sized balled nurserystock of Acacia cynophylla is planted at least60 cm deep in the sand with only the uppertwo to three leavesemerging (Mayerson, 1961).Costin et al. (1974) obtained better resultsby deep planting (up to 150 cm down) oflong cuttings of Tamarix aphylla, havingdiameters not less than 1.5 to 2.0 cm in theupper part of the dune. They found thatit was economical to resort to shallower (50to 70 cm) planting in depressions, with deepplanting on high sites, e.g., 80 cm to 120cm on slopes and 150 cm on high dunes.

In the interdune plains where hard panexists close to the surface, it is necessaryto break the pan to a depth of at least 45cm to 60 cm by a tractor auger for betterseedling establishment and growth.

Espacemellt: The fluctuations in the mois-ture regime of sand dune at different sites,the water consumption of plant species, theirroot system characteristics, and purpose ofplantation should be taken into considerationwhile determining the spacing of plants. Asa rule of thumb, the lower the rainfall andlonger the dry period, the wider the spacingto be adopted to minimize competition formoisture and to reduce development of highmoisture stress in plants. In the case ofAcaciatortilis and other tree species, a spacing of4 x 2 m has been adopted, while in the IGNPStage II area, the spacing is 3 x 3 m. Inother areas of Rajasthan, a spacing of 5 x4 m is adopted. It has, however, been feltthat a spacing of 4 x 4 m·would be more

SAND DUNE STABILIZATION 235

I "

appropriate for IGNP Stage II areas (Soni,pers. com.). Observations have, however,revealed that Acacia tortilis, in spite of ir-rigation for four years, has not put on com-mensurate growth. It is, therefore, suggestedthat instead of raising pure plantation of A.tortilis, it should be planted wide apart andin the intervening spaces Calligollllntpolygolloides and Lasillnts sindiclls may beplanted, as these are the dominant components

I of the natural vegetation of the area. Thesespecies are equally eff~tive in controllingsand drift and in providing fodder and fuel-wood. In interdune areas, a spacing of 5x 5 m is usually adopted. In the areas withannual rainfall of 100 to 150 mm, a widerspacing needs to be adopted.

Watering: The scarcity of water maymake raising of irrigated plantations difficult.In exceptionally dry years, life-savingirrigationof seedlings becomes necessary for survivaland for the development of sufficiently deeproot system to tap available soil moisture.The sand should be deeply moistened toprevent the development of superficial rootswhich would later die. No information is avail-able on irrigation water requirement oftransplanted tree seedlings except for neemseedlings under Jodhpur conditions (Burmanet al., 1991). Its importance isgenerally realisedand such watering has even been practicedin Haryana and Rajasthan, by adopting ar-bitrary norms. For example, in the regionreceiving 248 to 393 mm annual rainfall inHaryanai three waterings at the rate of 16L planf , one at the time of planting, theTable 6. Seedling growth perfonnallce ill relatioll to site

other in October, and the third during Decem-ber-February, are given. Even though aprovision for two waterings of plants in thesecond year and one in the third year hasbeen made, these irrigations are generally notrequired if timely planting of well developedhealthy seedings is done. This amount of watermakes the sand moist down to a depth ofabout 25 cm. A thin layer of dry sand shouldbe spread around the seedlings after eachwatering, as it acts as mulch and reducesevaporation losses.

It is felt that until the research resultsare obtained relating to actual wateravailability in the field, such as how muchwater the trees really need, when and forhow long it is needed, and how long a treecan survive without irrigation in differentmonths of the year, several plant water-stressindicators can be used as a general guideto irrigation requirement, viz., darkening ofleaf colour in some species, curling or changesof angle of leaves, and reduction of stomatalaperture (Armitage, 1985).

It has been observed that growth of plantedseedlings on the windward slope of the dunevaries in relation to site. This variation isshown in Table 6. The variatiQn in seedlinggrowth is possibly due to variation in soilmoisture content at the four sites. The variationin moisture content at different sites is dueto the nature and thickness of the sand andthe size of sand grain (Singh and Shankar-narayan, 1986).

Growth Sites on \\indward slope of dunet/3 from heel upward Mid t/3 up to the crest Interdune flats

I " III IVExcellent X XGood X XPoor X X

236

--- .---

KAUL

In the interdune plains, the presence ofhard pan near the surface contributes to poorperformance of seedlings. Deep soil working,as mentioned earlier, will considerably im-prove seedling performance.·

Inter-culture operations: In the initialstages, when the root growth of the plantis limited, there will be competition for mois-ture with weeds. Therefore, in Haryana, threeweedings and hoeing in' the first, two in thesecond and one in the third year, have beenprescribed. If the plantation hls been properlyraised, no weeding and hoeing is generallyrequired .in the second and the third yearof planting. If other vegetation is vigorousduring the second year, it becomes necessaryto undertake weeding. In Rajasthan, however,no weeding or hoeing has been prescribed.

Casualty replacement: As a normal prac-tice, a provision of 20% is made for casualtyreplacement. It has, however, been observedthat replacement planting in the second year,even if it survives, does not put on growthand tends to stagnate. All efforts should, there-fore, be made to achieve as high a percentageof seedling establishment as possible in theyear of planting itself. This should be possibleparticularly in the IGNP Stage II areas whereprovision of copious irrigation has been made.The calendar of activities for planting seedlingsand grass slips is given in Fig. 2.

Aerial seeding

The Indira Gandhi Canal passes throughvast stretche~ of parched (100 to 150 mmrainfall) dune country. Due to sand-ladenwind, this prestigious canal is getting coveredby invading sands. The less accessible dunetrack with low moisture is so large that itwill take too long to afforest it by the con-ventional methods. With the assistance of theNational Wastelands Development Board,Ministry of Environment and Forests, aerialseeding was, therefore, undertaken in July

1987at two sites havingaverage annual rainfallof 250 mm and summer temperature 4rc.Of these two sites, 300 ha was taken nearvillage Sardarpur and 400 ha near villageMotigarh, located 50 to 60 km north-westof Bikaner on Bikaner-Anoopgarh road. Boththe areas were fenced. Dune cre::;ts andwindward slopes were covered with 3 x 3m checker board mulch, which was preparedby burying the Crotalaria burhia bushes upsidedown.

Since the aim was to provide a silvi-pasturesystem, a mixture of four tree species, viz.,Acacia tortilis, Colophos.nenllll1ll mopane,Prosopis cineraria, Ziziphus rotlllldifolia, oneshrub, viz.,Dichrostacl'Ys nutans, one creeper,viz., Citnt/lllS colocynthis and one palatablegrass, viz., LasillntS sindiclls, were selected.A seed rate of 14 kg ha,l for the seed mixtureof tree, "hrub and grass was used. Tree andshrub seeds were treated with dilute sulphuricacid, washed in running water, and soakedin aldrin. LasillntS sindicllS seeds were pel,letised in mixture of clay, sand and ccwdung.

Based on evaluation of the two-year data,it was concluded tltat afforestation of inac-cessible dunes and sandywastelands ispossiblewith aerial seeding technology, provided seedsof suitable species in appropriate mixture arebroadcast in monsoon, adopting all pest con-trol measures (Shankarnarayan and Kumar,1988). Chances of success will no doubt in-crease if the aerial seeding happens to coincidewith a good rainfall year. Early attempts ofaerial seeding at Gadra Road in Barmer dis-trict (100 mm rainfall zone) were, however,not successful because of the unfavourableclimate, unpelletised seed used, and no treat-ment for repelling pests or predators(Bhimaya, 1977).

Higher seed germination, better seedlingsurvival and excellent vigour were recordedon bare dune crests which were aerially seededin July 1987. This indicates that the direct

SAND DUNE STABILIZATION

0z0Vl

<:- ..,:::e<::::e~

0Z0

3 Vl

~ <: '.=

:::e<::::e~

0z0Vl

<:

:::e<::::e~

0z0Vl

<: ••!l cc .., j.g c ..,.• 0•... O€ :::e.E •• <:3 :::e~ ~

237

uc"'0Co••c:e:;:::e

'",!j"e••co••·r'3 gell:::

••c'Cu

~1:u

"crI;i.0

"Vl

Fig. 2. Calentiar of aClit'ilics for plamillg Irec s('cdlillgs alld grass slips 011 shijiill;;salld dUlles alld i/llcrdllllal plaills.

238 KAUL

- --- ------ --

seeding from aircraft is successful on baredunes which are in fact adding to and ac-celerating the menace of wind erosion. Thisis so because bare dunes maintain a favourablewater balance even long after the rains areover (Gupta, 1979). Further, there is nocompetition for moisture from other vegeta-tion.

Economics of Dune Stabilization

Cost

It is difficult to give average figures forthe cost of sand dune fixation through af-forestation as expenses are likely to vary fromregion to region.

The per ha cost of sand dune afforestationfor a period of three years in Haryana worksout to Rs. 7,770 and for Rajasthan, Rs. 6,810.In the case of sand dune fixation in the IGNPStage II area, the cost per ha for a periodof eight years was worked out as Rs. 16,127.

Productioll

The absolute value of plantations estab-lished to protect people, habitations, otherstructures, roads, fields, or canals from in-vading sand, is often difficult to asses. How-ever, recent studies by Upadhyaya (1991) haverevealed that one and two-year- old plantationsreduced sand deposition by 0.513 m3 and1.023 m3 per running metre length of thecanal, respectively. In economic terms, theseplantations saved cost of desilting by Rs. 6,156per km in one-year-old plantations andRs. 12,276 per km in two-year-old plan-tations. The following information on firewoodand fodder yield from treated dunes givesa fair idea of their production potential.

Bhimaya et al. (1961) reported that atthe end of fifth year of afforestation, shiftingdunes produced 15 and 20 t ha'l wood inregions receiving 200 and 360 mm rainfallper annum, respectively. Studies by Kaul andGanguli (1964) revealed that sand dunes af-

forested withProsopis ju/iflora, in areas receiv-ing annual rainfall of 150to 250mm, produced15 t ha'l of firewood at the end of the fifthyear. Bhimaya et al. (1967) reported widevariations in fuel wood yield with respect toage of the trees and habitat. Differencesin fuel wood yield between the habitatsgenerally followed the pattern of rainfall.

The performance ofAcacia torlilis on sanddunes has been very encouraging. At Barmer(280 mm rainfall), the seven-year-old treeshad attained an average height of 7 m andthe average expected yield at the end of 10years was in the order of 25 t ha.1. Muthana(1980) reported a fuel wood yield of 30 tha,l from A. torlilis planted on sand dunes.In Haryana, A. tortilis plantations haverecorded a mean annual increment of 3.0to 3.5 m3 ha'l on stabilized dunes and about4.5 m3 ha'i on shifting dunes (Jakati, pers.comm.). Fuel wood yield from stems andbranches of Calligollul7l polygolloides, whichgrows extensively on sand dunes and reachesthe size of a small tree in seven years, rangesfrom 13 to 28 kg tree'\ with an average of19 kg tree'l (Kaul, 1965).

In the case of perennial grasses, the op-timum forage yield in well established plotsof Lasiunts Silldiclls and Cellclmls sp. wasrecorded up to 3.6 and 4.7 t ha'\ respectively(Ahuja, 1977).

Research Needs

Although the techniques developed to sta-bilize sand dunes through biological methodshave been successful, a few questions stillremain to be answered in order to furtherrefine the technology.

Seedling survival and subsequent growthlargely depend on the availability of soil mois-ture. Optimum spacing for different speciesof trees, shrubs and grasses at the initial andsubsequent stages of growth is, therefore, tobe determined in relation to optimal utilization

SAND DVNE SfABILIZATION 239

of the available soil moisture. In this context,thinning schedule may go a long way for sus-tainable vegetal growth.

At present, five to nine-month-old see-dlings are being planted. Feasibility of plantingtall transplants of different tree species needto be explored in relation to seedling survivaland subsequent growth, their capability togrow without mulching, operations necessaryfor survival and subsequent growth of smallsize seedlings, and the cost effectiveness ofplanting such tall transplants need to bestudied.

Relatively little comparative work has beendone in relation to irrigated tree crops. Thecurrent practice seems to be based on trialand error or on subjective experience andjudgement. Standardization of irrigationschedule for areas where water is available,is an important aspect of study in decidingthe optimal irrigation, amount of water ateach application, irrigation efficiency andseasonal water use.

Development of suitable silvipasture/silvi-horticulture models for shifting sand dunesin different rainfall regions is essential sothat information on species compatibility,geometry of planting and management prac-tices can be made available to farmers forapplication.

Potential productivity in relation to them<;\ny-sided protective and productive func-tions of different dune types, when afforested,need to be worked out for developingeconomic criteria.

References

Armitage, F.D. 1985. Irrigated Forestry in Arid and Semi-arid Lallds: A Syllthesis. International DevelopmentResearch Center, Ottawa, Canada. 160 p.

Den Salem, D. 1985. Mulching technique of dune fIXation:The layer system. FAO COlISen'atioll Guide, pp.71-74.

Bhimaya, CP. 1977. Ecosystem in arid zones. In DesertEcosystem and its Improvement (Ed. H.S. Mann),Jodhpur. CAZRI Monograph No. I, pp. 178-192.

Bhimaya, CP., Jain M.B., Kaul, R.N. and Ganguli, B.N.1967. A study of age and habitat differences infuel yield of Prosopis juliflora. Indian Forester93(6): 355·358.

Bhimaya. CP .. Kaul, R.N. and Ganguli, D.N. 1961. Sanddune rehabilitation in western Rajasthan. Proceed·illgs of the 5th World Forestry Congress, Vol 1,pp. 358·363.

Burman, V., Kathju,S., Garg, B.K and Lahiri, AN.1991.Water management of transplanted seedlingsof Azadirachta indica in arid areas. Forest Ecologl'alld Mallagemelll 40: 51-63.

Costin, E., Baleidi, AS., Ezz Mohamed and Dazare,M. 1974. Sand dune fixation and afforestation insome semi-desert countries of the Middle Eastarea. In Reporr on the FA O/DAN IDA Illter-RegionalTrainillgoll Heathlalld and Salld DuneAfforestatioll.FAO. Rome, pp. 66-69.

Dhir, R.P. 1985. Characteristics and properties of duneand associated soils. In Sand Dune StabilizatiOl~Shelterbelt alld Afforestation in Dry Zones, FAO,Rome, pp. 41-49.

Dhir, R.P., Kolarkar, AS., Sharma, B.K and Sharma,M.L. 1974. A note on particle size distributionin surface crust of an arid zone soil. JOLimal ofIndian Society of Soil Science 22: 377- 378.

Gupta, J.P. 1976.Studies on sand dunes. Annual ProgressReporr. CAZRI, 240 p.

Gupta, J.P. and Aggarwal, R.K 1980. Sand movementstudies under different landuse conditions ofwestern Rajasthan. In Arid ZOlle Research alldDel·c!opmellt. Scientific Publishers, Jodhpur, pp.109-114.

Kaplan. J., Karschon, R. and Kolar, M. 1970. Israel.In Afforestation in Arid Zones. (Ed. R.N. Kaul).Dr. W.Junk N.V. Publisher, The Hague, pp.136-154.

Kar, A 1993. Aeolian processes and bedforms in theThar desert. Joumal of Arid Environments 25:83-96.

Kaul, R.N. 1965. Silvical characteristics of Ca/ligonwnpo~\'gonoides Linn. La-Yaarall 15: 1-8.

Kaul, R.N. 1970. Indo-Pakistan. In Afforestation ill AridZOlles (Ed. R.N. Kal.I). Dr. W. Junk N.V. Publisher,The Hague, pp. 155- 209.

Kaul, R.N. 1983. Some silvical aspects of sand duneafforestation. 171e Imematiollal Tree Crop JOLimal2: 133-146.

Kaul, R.N. 1985a. Sand dune fixation and afforestationtraditional procedures for dune fIXation:The hedgesystem. In Salld Dune Stahilization, Shellerbelts alldAfforestatioll ill Df)' ZOlles. FAO Consen'atioil Guide,pp. 65·74.

240 KALIL

Kaul, R.N. 1985b. Afforestation of dune areas. In SandDune Stabilization, Shelterbelts and Afforestation inDry Zones. FAO Conservation Guide, pp. 75-85.

Kaul, R.N. 1992. Greening of Sand Dunes and InterdunalPlains. NAEB, Minist!)' of Environment & Forests,Govt. of India, New Delhi, 95 p.

Krishnan, A. 1977. A climate analysis 0; the arid zoneof India. In Desertification and its Control. ICAR,New Delhi, pp. 42-57.

Krishnan, A, Bhatt, P.N. and Rakhecha, P.A. 1966.Soil moisture regime and microclimatological studyover sand dunes in western Rajasthan. Annalsof Arid Zone 5: 1-9.

Lahiri, AN. 1964. Moisture relation of some desertplants. In Proceedings of the Symposium onProblems of Indian Arid Zone. LINESCO and Min-ist!) of Education. Government of India, 154 p.

Lahiri, AN. 1977. Plants in relation to soil moistureconditions of arid areas. In Desertification and itsControl, ICAR, New Delhi, pp. 225-236.

Mann, H.S., Lahiri, AN. and Pareek. O.P. 1975. Astudy on the moisture availability and other con-ditions on unstabilized dunes in the context ofpresent land use and the future prospects of deser-tification. Proceedings of the Indian NationalScience Academy 45B: 1-/7.

Mohammad, N. 1989. Rangeland Managemcllt in Pakis-tan. ICIMOD, Senior Fellowship Series No.1,ICIMOD, Nepal, pp. 1-193.

Pandey. S., Ghose, B. and Singh. S. 1964. Orientation,distribution and origin of sand dunes in the central

Luni basin. In Proceedings,Symposium on Problansof Indian Arid Zone. Govt. of India and' UNESCO,CAZRI, Jodhpur, pp. 84· 91.

Saxena, S.K. 1977. Vegetation and its succession in theIndian Desert. In Desertification and its Control.ICAR, New Delhi, pp. 176-192.

Shankarnarayan, K.A and Kumar, S. 1988.Aerial seedingof s.~nddunes. In Wastelands DevelopmClltand theirUtilization (Ed. K.A Shankarnarayan). ScientificPublishers, Jodhpur, pp. 237-256.

Sheikh, M.I. 1986. Afforestation of Arid and SClni-aridAreas In Pakistan. Pakistan Forest Instituie,Peshawar, pp. 1-154.

Singh, S. and Shankarnarayan, K.A 1986. Morphologyand management of sand dunes in the great Indiandesert. CurrelllPracticesin GCOlecllllicalEngineering3: 2347.

Upad:layaya, AK. 1991.Shelterbelt plantation effectivelycheck sand deposition in Indira Gandhi Canal. In-dian Forester 117: 511-514.

Venkateswarlu, B. and Ran, AV. 1981. Distributionof microorganisms in stabilized and unstabilizedsand dunes of the Indian desert. Joumal of AridEnl'irolllnents 4: 203-207.

Walker, 1\.S. 1982. Desert of China. American SciClltist70: 366- 376.

Zibu, Z.D. and Liu, S. 1979. nle Prel'elllion and Controlof Deserts and Desertification in China. LazhonInstitute of Desert Research.