Abstract: Landslide hazard zonation in and around Thodupuzha – Idukki – Munnar road (TM Road) in Idukki district,Kerala, India has been carried out using geospatial techniques. Being a landslide prone area a hazard zonation is attemptedusing terrain fragility concept. Based on the traverse mapping, slide prone areas and palaeo-slides along the TM roadwere identified. Precambrian crystallines consisting of hornblende-biotite gneiss, biotite gneiss, granite gneiss, charnockiteand pink granites form the main rock types. Factor maps of various terrain parameters such as slope, landuse, relativerelief, drainage pattern, drainage density, landform, and surface material were prepared and their integration carried outon a GIS platform. Based on geospatial analyses, the study area (438 sq. km) is ranked into four classes of relativefragility viz. highly fragile (8.25 sq. km), fragile (41.25 sq. km), moderately fragile (232 sq. km) and stable (156.5 sq.km). The first two categories together form 11% of the area, the most hazardous regions, which require immediatemitigation measures for slope protection. The study forms a basis for evolving a strategy for the development of theentire TM road of Idukki district. The fragility concept used in this study is a fast and cost effective model for identifyinglandslide prone areas, especially in the Western Ghats.
Keywords: Slope stability, Fragility, Landslides, Geospatial analyses, Western Ghats, Kerala.
et al., 1995; Thampi et al., 1997, Thampi 2006, 2009 andSajinkumar et al., 2011, Biju Abraham, 2011). Studies bySeshagiri et al., 1982, Thampi et al., 1997, Thampi 2006and Sreekumar et al, 2010, Sreekumar and Arish Aslam(2011) have proved that the slope failures in Western Ghatsare generally confined to the overburden. Krishnanath etal., 1996, Sreekumar and Krishnanath 2000, have studiednumber of profiles within Idukki and Pathanamthitta districtsand identified a number of profiles which are at thegeotechnical threshold.
STUDY AREA
The study area (Fig. 1) forms part of the Western Ghatsin Idukki district, Kerala State. The study area is locatedaround Thodupuzha- Idukki- Adimali -Munnar road (TM -road) within latitude 9° 46' & 10° 05’N and longitude 76°45' & 77° 05' E falling in the Survey of India topographicsheet Nos. 58 C/13, 58 C/9, 58 F/4 (Fig 1).
GEOLOGICAL SETTING
A geological map of the entire study area (Fig. 2) isprepared using topographic sheets (scale 1:50,000), regional
INTRODUCTION
Landslides are a form of mass movement, which can bedescribed as a rapid down slope movement of soil, sedimentsand rocks due to gravity (Coates, 1981). It characteristicallyoccurs along one or more discrete bounding slip surfaces(Hutchinson, 1988). Geomorphological features, slopes,climate and human intervention play an important role intriggering the process. Every year with the onset of monsoonreports pour in from almost all the highland districts,especially along the slopes and communication networksin Idukki district about ‘slope failures’/ ‘landslides’. Thisphenomenon is locally known as ‘Urulpottal’.
These recurring disastrous events in and aroundThodupuzha- Idukki- Adimali -Munnar road (TM - road),Idukki district warrant a detailed study to identify the fragileareas in terms of slope stability. Among all the naturalhazards, slope failures are most amenable to correctivemeasures aimed at prevention.
Several workers have investigated slope failures inWestern Ghats and occurrences of landslides in the Idukkidistrict (Seshagiri et al., 1982, Kandaswamy et al., 1985Krishnanath.et al., 1985, 1993, 1996, Biju Abraham.et.al.,Krishnanath and Sreekumar, 1996, Sreekumar, 1998,Sreekumar and Krishnanath, 2000, Sankar, 1991; Earnest
JOUR.GEOL.SOC.INDIA, VOL.82, DEC. 2013
650 P. BIJU ABRAHAM AND E. SHAJI
geological maps of Geological Survey of India and fromthe data generated in this study. The slope forming materialexposed along the road cuttings consists predominantly ofcrystalline rocks, and their highly weathered equivalents.The main rock types include biotite gneiss (granitic gneiss),hornblende biotite gneiss (composite gneiss), granite gneiss,charnockites and pyroxene granulite. These rocks areintruded by basic, acidic and alkaline intrusives (Fig .2).Pink granites are seen in the Munnar area
The general trend of foliation is NW-SE with deviationsranging from NNW-SSE to NNE-SSW (Fig. 2). The foliationdip is steep to vertical (50° to 90°), low dips being rare. Allthese rocks are capped with varying thickness of weatheredmaterial. The thickness of the weathered material varies from1 to 23 meter and is site specific. These weathered materialsare less coherent and structurally anisotropic, which thusbecomes one of the dominant lithology exposed.
MATERIALS AND METHODS
A base map of the study area has been prepared fromthe survey of India toposheets (scale 1:50,000) with field
survey, to locate the palaeo as well as potential landslideoccurrences in the study area. Detailed geological mappingof various profiles was carried out along the TM-road foridentifying potentially unstable profiles.
Geospatial analysis is attempted using various thematicmaps prepared on a GIS platform (scale 1:50,000) usingtopo-sheets, remote sensing and field data. The methodologyinvolves the preparation of various thematic/ factor mapsusing GIS software. The toposheets (scale 1:50,000 and1:25,000) were scanned and geo-referenced using ArcGIS9.1 version. The geo-referenced maps are then digitized toconvert the data from raster format to vector format. Vectordata are stored in ArcGIS as various themes in the shapeformat. The line features are then converted to polygons, tocompute the area of the polygons. The digitization, editingand cleaning operations results in the creation of variousthematic maps. The theme and the associated legend definethe symbolization of features. The methodology involvesallotting weights or marks to these causative factors andranks the terrain as per the total weight into differentcategories of fragility. A summation of all parametersgenerated through terrain evaluation is used to estimate thefragility of the terrain. The study area was divided into squaregrids each of dimension 1 cm, 500 m x 500 m or 0.25 sq.km(total 1752 grids for an area of 438 sq.km). This grid issuperposed on each factor map and respective weight ofFig.1. The study area with important locations along Thodupuzha–
Idukki–Adimali-Munnar road.
Fig.2. Geologic map of the study area (modified after GSI, 1989).
JOUR.GEOL.SOC.INDIA, VOL.82, DEC. 2013
LANDSLIDE HAZARD ZONATION IN AND AROUND THODUPUZHA–IDUKKI–MUNNAR ROAD, KERALA 651
each factor is allotted for all grids. After repeating theexercise with all the factor maps the cumulative weight valuefor each grid of the study area was arrived at. Based on thecumulative weight value of all factors the grids arecategorized into highly fragile, fragile, moderately fragileand stable, resulting in the preparation of a fragility(zonation) map. In addition to getting an overall fragilitypicture of the segment the output will give the individualhot spots as far as terrain stability is concerned on 1:50,000scale. The overall terrain fragility (FI –fragility index) of asegment is calculated as follows
FI = W /G,
where, FI – fragility index; G - is the total number of gridsfalling within the segment; W - is total weight of all gridsfalling within the segment.
Terrain Evaluation using Geospatial Analysis
For a given area, depending upon the prevalent failureinducing factors along with their severity and interactionpossibility, the degree of fragility could also be worked out.Fragility assessments can be used to predict the geographicallocation of terrain hotspots. Even though this process canbe applied to delineate critical areas in landslide hazardzonation, it does not forecast “when” or “how frequently” alandslide will occur or how large and destructive the slopefailure will be. The combined influences of slope, landuse,relative relief, drainage pattern, drainage density, landform,surface/cover material, geology, geotechnical properties etc.are evaluated in slope stability studies.
As such, in assessing the fragility of the study area themain parameters which collectively influence the fragilityof the overburden is only considered. The followingparameters were studied and thematic maps prepared forthe fragility estimation study: (1) slope; (2) landuse;(3) relative relief; (4) drainage pattern; (5) drainage density;(6) landform and (7) surface material.
Appropriate weightage is allotted to the aboveparameters based on the terrain conditions, field data,methodology adopted by Center for Earth Science Studies,
Trivandrum in preparing district level multi hazard maps ofKerala (CESS, 2010) and weights are given in the Bureauof Indian Standards (IS 14496, 1998). This procedure isadopted by many workers based on terrain conditions(Thampi, 2006 and John Mathai, 2009). Weights allottedto each parameter are given in the Table 1.
All the parameters (Table 1) were further sub-divided tocalculate the fragility of the terrain at micro level. Detailsof the individual terrain parameters are discussed below
Slope
The stability of a slope segment is dependent on variousfactors like slope angle, slope form, slope length, materialof which it is formed, antecedent moisture content etc.
The angle of slope steepness is calculated using (Young’smethod) the basic trigonometric formula, TanU = V/d, where,U is the angle of slope, V is the vertical contour interval(altitude difference between two points in the same line ofthe slope) and d is the contour separation (horizontal distancebetween the two points). Slope map of the terrain is prepared(Fig.3) with digitized 20 meter contours on a GIS platformusing TIN (Triangulated Irregular Network). These slopesare then grouped in to four classes (Table 2) so as toincorporate all the predominant slopes of the area. The
Table 1. Weight allotted for the different terrain parameters
Sl. Terrain WeightNo Parameter Allotted
1 Slope 302 Landuse 203 Relative relief 104 Drainage pattern 105 Drainage density 106 Landform 107 Surface material 10
Total Weight 100
Table 2. Weights allotted, area and percentage distribution of slopes
Slope Weight Area in Percentage ofMagnitude sq. km aerial occurrence
following weights are allotted for each category of slopes.Even though the slope greater than 50o is given maximumweightage, the area having slope between 16o – 35o isidentified as critical.
Landuse
When considering the landuse of the study area, the maindeterminant factor that comes up is the vegetation coversince the major part is covered by plantations and naturalvegetation. Detailed studies in the Western Ghats regionindicate that the highest incidence of slope failures are seenwithin degraded forest areas in the upper slope regions andin waste land without vegetation. The discontinuous soilcovered zones among rock outcrops forming the plateauedges with natural degraded forests and grasslands are zonesof high susceptibility. The undisturbed forest, natural grasslands and forest plantations have not indicated the incidenceof slope failures. Tea plantations generally are not
JOUR.GEOL.SOC.INDIA, VOL.82, DEC. 2013
652 P. BIJU ABRAHAM AND E. SHAJI
susceptible to land disturbances. Other plantations such ascoffee, cardamom, rubber, etc show incidence of slopefailures mainly due to improper land management practicesand cultivation of seasonal crops in vulnerable zones. Incase of plantations the most susceptible time for slopefailures is when the older plants are removed for replantation.
Landuse map of the study area was prepared based on1:50,000 and 1:25,000 toposheets, IRS satellite data, quickbird image (www.googleearth.com), with limited fieldchecking. The landuse map of the terrain delineating variouslanduse categories (Fig. 4) shows all the landuse patternsbased on the signature of the vegetation. Landuse and theiraerial distribution are given in Table 3. Six major landusepatterns are identified and appropriate weights are allottedfor each category of landuse as shown in the Table 3. Thedegraded forest and grasslands covering an area of 135sq.km are identified as more susceptible to slope failures.
Relative Relief
The relative relief of a terrain reflects thegeomorphological history of the area. Relative reliefrepresents actual variation of altitude in a unit area withrespect to its local base level (Kumar and Pandey, 1982).Relative relief being the difference in height between thehighest and lowest points per unit area and is directly relatedto degree of dissection (Saxena & Prakash, 1982). The morethe intensity of dissection, greater is the relative relief. Hencethis parameter finds relevance in slope failure studies. Therelative relief map of the study area was compiled using aunit area of 500m x 500m (Fig. 5). The relevant data isrepresented as a contour diagram. The following weightsare allotted for each category of relative relief. Table 4give the details of weights allotted, area and percentagedistribution of relative relief of the study area. About120 sq.km area is identified as unstable based on therelative relief data.
Fig.3. Slope map of the study area.
Table 3. Weights allotted for Landuse
Landuse category Weight Areain sq.km
Waste land without vegetation 20 84Mixed crops/seasonal 15 76Built up land 10 22Degraded Forest 10 112Grassland , scrub jungle & rock with grass 5 23Forest(closed) and plantations 0 121
Table 4. Weights allotted, area and percentage distribution of relativerelief
Relative relief Weight Area in Percentage ofin meters sq.km aerial occurrence
< 30 0 28.1 6.42
30–100 5 290.4 66.3
> 100 10 119.5 27.28
Fig.4. Landuse map of the study area.
JOUR.GEOL.SOC.INDIA, VOL.82, DEC. 2013
LANDSLIDE HAZARD ZONATION IN AND AROUND THODUPUZHA–IDUKKI–MUNNAR ROAD, KERALA 653
Drainage Pattern
From the point of view of slope stability the paralleldrainage pattern is very critical as it indicates high slopes.Further, this pattern generally comprises of first orderstreams forming the source region of a drainage basin. Suchareas are zones of considerable erosion and slope retreat.Figure 6 shows the drainage pattern map of the area. Assuch in the map prepared only parallel pattern is shown andthe rest are clubbed together. Table 5 gives the details ofweights allotted, area and percentage distribution of drainagepattern of the study area. Based on the drainage patternanalysis, an area of 100 sq.km is identified as critical.
weights allotted, area and percentage distribution of drainagedensity of the study area. About 100 sq.km area has highdrainage density.
Table 5. Weights allotted, area and percentage distribution of drainagepattern
Drainage category Weight Area in Percentage ofsq.km aerial distribution
Parallel 10 108.9 24.87
Others 0 329.91 75.32
Fig.5. Relative relief map of the study area.
Table 7. Weights allotted, area and percentage distribution of landform
Landform Weight Area in Percentage ofsq.km aerial distribution
Table 6. Weights allotted, area and percentage distribution of drainagedensity
Drainage density( length per 1 cm grid) Weight
>1 Km 10
0.5 – 1 Km 5
<0.5 0
Fig.6. Drainage Pattern map of the study area.
Drainage Density
High drainage densities are indicative of imperviousstrata, high rainfall, less vegetation and active stream incisionall of which may be associated with mass movements. Thedrainage density map (Fig. 7) is prepared and weights areallotted for various categories. Table 6 gives the details of
Landform
The landform map (Fig. 8) of the area is prepared andweights are allotted for various categories. Table 7 givesthe details of weights allotted, area and percentagedistribution of landform of the study area. The area beingthe upper catchment of river basins show youthful fluviallandform signatures. Elevated gently undulating crustalportions are classified as hill crests. The hill crests are
JOUR.GEOL.SOC.INDIA, VOL.82, DEC. 2013
654 P. BIJU ABRAHAM AND E. SHAJI
terrain with upper slope (125 sq. km) is found to bevulnerable if the rainfall intensity exceeds 195 mm.
Surface Material
Surface material map is prepared (Fig. 9) and identifiedvarious categories of cover material and weights allottedfor different categories. Table 8 gives the details of weightsallotted, area and percentage distribution of surface materialof the study area. The surface material or overburdenmaterial accumulated over a period time on the freshbasement rock plays a vital role in the fragility of slopes.Depending upon their physical properties the water holdingcapacity of the slope varies. An important stability aspectnoted in this region is that normally the most prevalent typesof slope failures are confined to this over burden only withoutaffecting the basement rocks. The area covered with debrismantle (215 sq. km) with slope between 16o and 35o isidentified as most vulnerable.
Fig.8. Landform map of the study area.
Fig.7. Drainage density map of the study area.
Table 8. Weights allotted, area and percentage distribution of Surfacematerial
Surface material Weight Area Percentagein sq.km of aerial
flanked by moderate to steeply sloping side slopes. The sideslopes merge with the concave foot slopes with a break inslope. Foot slopes are not developed everywhere but whendeveloped they merge with the narrow valley floor. The
DISCUSSION
The summation of the seven parameters generatedthrough terrain evaluation gives the required clues on thefragility of the terrain. The overall terrain fragility of asegment is calculated using the formula given in themethodology. The fragility classification is presented inTable 9. The entire study area is further divided into DivisionA (Lower area between Arakkulam and Painavu), DivisionB (Middle area between Painavu and Adimali) and DivisionC (Upper area between Adimali and Munnar). Based onthe cumulative weight value of all factors (Table 9) the gridsare categorized into highly fragile, fragile, moderately fragileand stable. The zonation map based on fragility is given inFig.10. The different colors in the fragility map define theareal extent and degree of vulnerability. When the totalweight of a grid is greater than 75, the grids are classified ashighly fragile zone. These are very unstable zones whereslope failures are likely to occur. Three highly fragilesegments along the TM road are identified, and located
JOUR.GEOL.SOC.INDIA, VOL.82, DEC. 2013
LANDSLIDE HAZARD ZONATION IN AND AROUND THODUPUZHA–IDUKKI–MUNNAR ROAD, KERALA 655
between Puchappara – Kulamavu sector (3 km), Panamkutti–Adimali sector (4 km) and Adimali- Munnar sector (2 km).When the total weight of a grid is between 61 and 75 it isclassified as fragile zone. Terrain setting almost comparable
to the first category but failure proneness is amenable tocost effective mitigation measures like slope correction,encouraging regeneration of natural vegetation,reforestation, drainage correction etc. When the total weightof a grid is between 40 and 60 then the grids are classifiedas moderately fragile zone. These areas are stable in thepresent state of existence. However, it is necessary that futurelanduse activity is to be properly planned to maintain thepresent status. When the total weight of a grid is less than40 it is classified as stable zone. These areas are stablewhere development activities are possible.
CONCLUSIONS
Concept of terrain fragility is introduced for the firsttime in the Idukki district for determining a fragility rankingin terms of landslides. The geospatial analysis and fragilityranking map provides a general picture on the stabilityscenario along the Thodupuzha – Idukki- Adimali – Munnarroad. The total study area of 438 sq.km is divided into 1752square grids to find out the hot spots. Geospatial analysesbased on the cumulative weight value of all factors rankedthe study area into four classes of relative fragility viz. highlyfragile (8.25 sq.km, 1.88%), fragile (41.25 sq.km, 9.42%),moderately fragile (232 sq.km, 52.97%) and stable (156.5sq.km, 35.73%). The first two categories form 49.5 sq.km
Fig.10. Fragility ranking of the study area.
Table 9. Fragility classification of the study area
Rank Category Total Area in Percentage No. ofWeight Sq.km distribution cells
or 11 % of the study area. From this analysis it is evidentthat the real hot spots form about 2% of the study area andare located between Puchappara – Kulamavu sector (3 km),Panamkutti – Adimali sector (4 km) and Adimali- Munnarsector (2 km) of the TM road. The highly fragile and fragileareas require immediate mitigative measures for slope
protection. Further developmental activities in this arealike road cutting, construction of new structures, terrain andlanduse modifications are to be done only after implementingproper mitigation strategies. As such this model is userfriendly and economic in making a total and all pervasiveterrain ecosystem assessment in the high land terrain.
References
BIJU ABRAHAM, P. (2011) Slope Processes and Stability AlongThodupuzha – Idukki – Adimali- Munnar Road, IdukkiDistrict, Kerala State, Unpublished Ph.D Thesis, Universityof Kerala, India.
BIJU ABRAHAM, P., KRISHNANATH, R. and Sreenath, G. (1996) SlopeStability Along Highways-A study around Karimpan,Chelachuvadu and Nagarampara areas in Idukki District,Kerala State, India. Proc. International Conference on Disastersand Mitigation, Anna University, Madras, A4-29-34.
CESS (2010) Plan Project 239. Detailed Project Report.COATES, D.R. (1981) Environmental Geology. John Wiley, New
York, 701pEARNEST, A., ARFKHAN and DEVPALAN NAIR, G. (1995) Role of
groundwater in landslide initiation on Western Ghats, Newcase histories from Palakkad district Kerala, Abstr. National.Seminar on landslides in Western Ghats, CESS, Trivandrum
HUTCHINSON, J.N. (1988) General Report: Morphological andgeotechnical parameters of landslides in relation to geologyand hydrogeology Proc. 5th Internat. Natl. Symp. Landslides,v.1, Balkema, Netherlands, pp.3-35.
IS 14496, part 2. (1998) Indian Standards for preparation oflandslide hazard zonation map in mountainous terrain – Guidelines, Macro Zonation – BIS 1998. Author: Dr. Bhavani Singh.
JOHN MATHAI (2009) District level natural hazard zonation mapsfor Kerala State, Proc. Nat. Workshop Natural Hazards. Centrefor Earth Science Studies.
KANDASWAMY, L.C., VASU, K. and VINAYAN, P.K. (1985) Probabledirections of investigation of Landslide with special referenceto Wayanad district, Western Ghats, Proc. Workshop onLandslides in Western Ghats, CWRDM, Calicut.
KRISHNANATH, R. and SREEKUMAR, S. (1996) Identification andlocation of slide prone areas in certain hard rock profiles ofWestern Ghats. Indian Geotechn. Jour., v.26, pp.407-416.
KRISHNANATH, R., BIJU ABRAHAM, P. and SREENATH, G. (1993) SlopeStability analysis at Kadamurutti profile in Thodupuzha-Idukkiroad, Idukki district, Kerala. India. Proc. Seminar on Land-slides in Kerala, College of Engineering Trivandrum, pp.1-7.
KRISHNANATH, R., NANDAKUMAR, P., DEVARAJAN, K. and JOSE JAMES.(1985) Geotechnical Factors Influencing Slope Stability - Casestudies of Lockhart and Devikulam Profiles ,Idukki District,Kerala, Workshop on Landslides in Western Ghats, Calicut,Proc. State Committee for Sci. Tech. and Environment,Trivandrum, pp.34-43.
KRISHNANATH, R., SREENATH, G. and BIJU ABRAHAM, P. (1996)Seepage Modeling and Slope Stability analysis at Devikulam– Manale profile along Kumili –Munnar road in Idukki district,Kerala. India. Jour. Environ. Pollution 3(3 and 4), TechnoSci. Publ., pp.215-220.
SAJINKUMAR, K.S., ANBAZHAGAN, S., PRADEEPKUMAR, A.P. and RANI,V.R. (2011) Weathering and landslide occurrences in parts of
Western Ghats, Kerala. Jour. Geol. Soc. India, v.78, pp.249-257.
SANKAR, S. (1991) Investigation of landslides in Kodaranji andKurachundu. Calicut district. Report. CESS, Trivandrum.
SAXENA, P.B. and PRAKASH, S. (1982) A study of the morphometricdeterminants of the stage of cycle of erosion in the Nayar Basin.(Garhwal Himalayas). In: H.S. Sharma (Ed.), Perspectives inGeomorphology. Concept Publ. Co., New Delhi, pp.77–92.
SESHAGIRI, D.N., BADRINARAYANAN, S., UPENDRAN, R., LAKSHMI-KANTHAM, C.B. and Srinivasan (1982) The Nilgiris landslide.Geol. Surv. India Misc. Publ., no.57.
SESHAGIRI, D.N., UPENDRAN, R., LAKSHMIKANTHAM, C.B. andSUBRAMANIAN, K. (1982) Landslide zonation in Nilgiri Plateau,Tamil Nadu, India. Proc. IV Congress Intl. Nat. Assoc. Engg.Geol., 1, 379 - 92.
SREEKUMAR, S. (1998) Geotechnical and Petrological studies ofslopes along Kottayam-Kumili road, Kerala state, India.Unpubld. Ph.D thesis, Department of Geology, University ofKerala, Trivandrum, India, 320p.
SREEKUMAR, S. (2009) Techniques for Slope Stability Analysis,Idukki Dist., Kerala. Jour. Geol. Soc. India, v.73, pp.814-820.
SREEKUMAR, S. and ARISH ASLAM (2011) Micro Zonation ofLandslide Hazards in Idukki district, Kerala. Proc. of NationalConference on Advances in Engineering Geology andGeotechnical Engineering, Govt. College of Engineering,Kannur, pp.32-38.
SREEKUMAR, S. and KRISHNANATH, R. (2000) Stability of Lateriticprofiles in parts of Western Ghats, India. In: E.N. Bromhead,N. Dixon, M-L. Ibsen and E. Sfakianaki (Eds.), LandslidesResearch: Theory and Practice: Proceedings of the 8thinternational symposium on landslides, Cardiff, England,Thomas Telford Ltd., London, v.3, pp.1395-1400.
SREEKUMAR, S., BIJU ABRAHAM, P. and SREENATH, G. (1995)Geometric analysis of certain hard rock profiles in Idukkidistrict, Kerala state. Abstr. National seminar on Landslidesin Western Ghats, CESS, Trivandrum.
SREEKUMAR, S., ROSHNI, K., AAARON WILSON and RATHEESH KUMAR,M.D. (2010) An Evaluation of Landuse Changes and LandDegradation- A case study from Eranad taluk, Malappuramdistrict, Kerala. In: Venkatachalapathy (Ed.), Earth Resourcesand Environment, Chapter 19.
THAMPI, P.K. (1997) Evaluation study in terms of landslidemitigation in parts of Western Ghats, Kerala. Report CESS,Trivandrum.
THAMPI, P.K. (2006) Landslide Management in Kerala. Compen-dium on Focal theme. XVIII Kerala Science Congress.
THAMPI, P. K., JOHN MATHAI., SANKAR, G. and SIDHARTHAN, S. (1997)Evaluation study in terms of landslide mitigation in parts ofWestern Ghats, Kerala. Ministry of agriculture, Govt. of India,Publ., Centre for Earth Science Studies, Trivandrum.
(Received: 1 June 2012; Revised form accepted: 11 November 2012)
