combining data from satellite images and reindeer herders in … · 2011. 4. 29. · and practical...

Nordia Geographical Publications 35: 2, 17–30

17

Timo Kumpula & al.

Abstract: The aim of the study is to assess the capacity for satellite imagery indetecting different natural and anthropogenic land cover types in the vicinity of amodern petroleum extraction development in the Russian Arctic. The Yamal Peninsulain northwest Siberia contains some of the largest untapped deposits known in theworld. It also serves as the homeland of the Yamal Nenets, who have exploited firstwild and then domestic reindeer in the region for at least 1000 years. Their annualmigration from the treeline to the northern tundra brings them into contact with anumber of impacts associated with gas exploration and production. These rangewidely and include physical obstructions from roads, railways, and pipelines, as wellas direct and indirect ecological impacts, such as changes in vegetation, soils andhydrology due to e.g. drilling, infrastructure development, and seismic surveys. Someof the effects are relatively small-scale, only a few meters across, while others coverseveral hectares. Nenets’ perceptions of the spatial aspects of their territoriesencompass changes in both quantity and quality of terrestrial habitats, rivers, lakesand campsites that have been used seasonally for centuries. Even high-resolutionimagery was unable to detect things like trash (rusted metal, broken glass), drillingmuds and petro-chemicals that can strongly affect the overall quality of reindeerpastures. To properly assess the overall ecological impacts of petroleum developmentrequires a combination of state-of-the-art remote sensing coupled with detailedground-truthing. These efforts must embrace both scientific and local knowledge fromindigenous herders and also non-indigenous gas field workers.Key words: Nenets, reindeer migration, indigenous knowing, remote sensing, arctictundra

Combining Data from Satellite Images and ReindeerHerders in Arctic Petroleum Development: the Caseof Yamal, West Siberia

Timo Kumpula1, Bruce Forbes2 & Florian Stammler2,3

1 Department of Geography, University of Joensuu2 Arctic Centre, University of Lapland3 Scott Polar Research Institute, University of Cambridge

Introduction

Russia is becoming more important as aproducer and supplier of oil and gas toEuropean and eventually North Americanmarkets. The Iraq war and post-war chaoshave partly increased world market price

of oil barrel to its record. Meanwhile Russiahas invested in pipelines across eastern andcentral Europe and tanker traffic via theNorthern Sea Route. Among the mainsources are the giant oil fields of the NenetsAutonomous Okrug (NAO), Khanti-MansiskAutonomous Okrug (KMAO), and thesuper giant gas fields of the Yamal-Nenets

NGP Yearbook 2006. Theme Issue on Northern Nature and Human Activities. Guest editor Olavi Heikkinen.Editors Katri Suorsa, Marja Anttonen & Marja Mönkkönen. Nordia Geographical Publications Vol. 35:2.ISBN 951-42-8326-0 ISSN 1238-2086. Publications of the Department of Geography, University ofOulu and the Geographical Society of Northern Finland, 2006. http://terra.oulu.fi/psms.htm

Combining Data from Satellite Images and... NGP Yearbook 2006

18

Autonomous Okrug (YNAO). Russianproduction has an image of being morestable and safe from political crisis thansome of the big middle Eastern producers.It has once again become a powerfultrading partner, in particular as a majorsource of energy for the global market. Atpresent Russia supplies 25% of the world’snatural gas, which makes it the No. 1producer. Some 90% of this productioncomes from West Siberia. As for oil, Russiarecently also took over the leading positionin world production from Saudi Arabia(Mosnews 2006). 60% of Russia’s oil isfrom West Siberia. Overall EU dependencyon both oil and gas from Russia is high andincreasing, relative to European sources likeNorway and the declining export capacityof the UK. In 2005 the European Unionreceived 25% of its oil (30% of all imports)and 25% of its gas (50% of all imports)from Russia. Dependency on Russia amongindividual EU countries, however, variesgreatly. For example, Finland relies 100%on Russian gas sources, whereas France withits traditional ties to Algeria imported only24% from Russia as of 2004.

As petroleum industrial infrastructureexpands rapidly throughout Russia’s north,it has profound implications for theenvironment and region’s economies andlocal people, in particular indigenous groupspracticing traditional livelihoods like herdingdomestic reindeer (Rangifer tarandus), huntingand fishing. The building up of the Russianempire has affected these indigenous groupsover the last 500 years. Up until the 19thcentury the main resource produced fromthese northern regions was fur, whereasfrom the mid-20th century onwardpetroleum products became the centralfocus.

The exploration and exploitation phasesof northern areas like YNAO and NAOwere launched in the 1960’s with extensiveresource prospecting. Geological surveyssearched through the tundra zone withheavy drilling equipment and left clearimprints on the permafrost landscapes.Surveys mapped the oil and gas deposits anddrilling was first started in boreal andsubarctic areas with easiest access from thesouth, such as existing railways and roads.Surveys and drilling further north in trulyarctic areas required enormous amounts ofinfrastructure for example building ofpipelines, expanded road and rail networksand accommodation facilities. Supply ofremote arctic exploration sites has beendifficult with winter road networks,helicopter, aeroplane and also by ships viathe Northern Sea route. In the post-Sovietperiod, oil and gas have become a vitalsource of profits to the overall Russianeconomy. The exploration and exploitationphases of petroleum development result inbenefits and wealth but also, at the sametime, in a large number of direct andcumulative impacts on ecosystems andcultures all along the route from the sourceto the market (Forbes 2004; Stammler &Wilson 2006). New drilling sites and oil andgas fields have been connected to transportnetworks to export oil and gas to worldmarkets. This of course has increased impactson the environment and local indigenouscommunities.

In the YNAO indigenous lifeways of theNenets, Khanty, Komi-izhemtsy and Sel’kupfishermen, reindeer herders and hunters - forthe many of whom are still nomadic or semi-nomadic – are comparatively well-preserved.While use of reindeer reaches back to atleast 1200 years, the large scale herding of


19

Timo Kumpula & al.

domestic reindeer has developed in the1600’s. In YNAO on Yamal Peninsulareindeer herding survived intact from theSoviet period with the least amount ofdamage compared to other indigenouscultures elsewhere in Russia (Stammler 2005).

Satellite images have been usedsuccessfully in the detection of marine andcoastal oil spills. Especially radar satelliteimages are suitable for detection of oil spills(Benelli & Garzelli 1999; Espedal &Johannessen 2000; Jones 2001; Solberg &al. 2003). Oil spill detection from terrestrialdrilling sites is more difficult, because thesize of spill is an important factor that definesits potential for detection. In detection ofimpacts not only spills should be detected.Impacts should be studied holistically, allpossible impacts on environment and otherland use forms. Also impacts on the herdingcommunity must be studied, and severalauthors have argued for a shift in the focusof impact studies from spill response andimpact and damage compensation toprevention of spills, minimising negativeimpacts and development plans for localpopulations (Meschtyb & al. 2005; Stammler& Wilson 2006).

Impacts of industrialisation in the northernRussian mining towns of Vorkuta andNoril’sk have been studied with Landsat TMdata (Toutoubalina & Rees 1999; Virtanen& al. 2002). Tømmervik & al. (2003) studiedeffects of air pollution on vegetation in theNorwegian-Russian border area by usingmultitemporal Landsat MSS/TM data.Stow & al. (2004) reviewed the possibilitiesof change detection studies in arctic areaswith available remote sensing data. Rees &al. (2003) studied land cover change in NAOwith Landsat TM and ETM+ images, theyalso attempted to incorporate indigenousknowledge of Nenets reindeer herders to

land cover change interpretation, althoughthis formed only a very small part of theiranalysis. One problem associated with suchattempts is that indigenous knowledge ishighly situative, very detailed, learned andenacted by people moving through theland. Prominent scholars have thereforequestioned the compatibility of “indigenousknowledge” with natural science knowledgethat can be put into databases and treatedas body of information independent fromits context, whereas indigenous ways ofknowing are culturally embedded sets ofpractices (Ingold 2005; Stammler 2005).Landsat MSS/TM/ETM+ data are toocoarse to detect small-scale impacts that arecommon in most cases. Very high resolutionimages like IKONOS-2 and Quickbird-2have a resolution of less than 4 meters andthey have the potential for very detailedimpact detection. In the Arctic there havebeen few studies where IKONOS-2 imageshave been used to study lichen pastures ofreindeer (Allard 2003; Kumpula 2006).

There is a critical need to minimizeconflicts between arctic grazing systems andindustrial development, particularly in Russiawhich remains in a state of drastic and rapidtransformation (Klein 2000; Krupnik2000). There is strong evidence that theseimpacts are reaching a critical threshold inYNAO, where the withdrawal of lands forindustrial infrastructure has pushed a steadilyincreasing number of animals ontoprogressively smaller areas of tundrapastures (Zen’ko 2004; Stammler 2005). Thisis leading to desertification in places and aserious decline in the quantity and quality ofthe remaining upland tundra suitable forreindeer pasture (Forbes 1999; Jernsletten& Klokov 2002). At the same time, theindigenous Nenets population has beenexposed to a number of new health and


20

demographic problems directly related tothe industrial development (Pika &Bogoyavlensky 1995).

In the project Environmetal and SocialImpacts of Industrial Development in NorthernRussia (ENSINOR) we focus on amultidisciplinary analysis of the social andenvironmental consequences of energydevelopment. We combine state-of-the-artWestern scientific technologies with theindigenous or local perceptions of processesof change by Nenets reindeer herders. Thismethod of co-producing knowledge canbe expected to yield results which are bothhighly relevant to local and regional needsand practical for policy implementation.The immediate objectives are to characterizethe effects of 2-3 decades of industrialdevelopment of petroleum on traditionallivelihoods and the supporting society inYNAO and NAO. In this article we arestudying the impacts that can be detectedfrom the different resolution satellite images.We also discuss the significance of fieldsurveys and local knowledge in enhancingthe interpretation of satellite images.

Research area

The Yamal-Nenets Autonomous Okrug(YNAO) belongs to the West Siberianeconomic region and the Ural Federaldistrict. YNAO is an independent unit ofthe Tyumen Oblast and lies in the extremenorth of the West Siberian lowland, whichencompasses the largest wetland (technicallya peatland) in the world. With a territory of750 300 km2, the length of YNAO fromnorth to south is 1200 km and from eastto west 1130 km. In the Nenets’ own

language, “Yamal” means “the end of theearth” or “land’s end”. Ice-rich permafrostis widespread and is susceptible tothermokarst erosion from both natural andanthropogenic disturbance. As a result, muchof the terrain is considered moderately toextremely unstable for purposes ofengineering and infrastructure developments(roads, bridges, pipelines, etc.). The forest-tundra transition zone covers a broad bandcoincident with the southern portions ofYamal and Taz Peninsulas and most mapsdepict the latitudinal treeline as runningapproximately through the city of Salekhardon the Arctic Circle (lat. 66°33’N). For theindigenous Nenets, their needs are metmainly by reindeer. South of the Ob Riverdelta, sparse lichen woodlands of Siberianlarch (Larix sibirica) give way to more orless closed forest in the southernmost partsof YNAO. On the shores of lakes, riversand bays, subsistence fishing is of majorimportance too. Substantial populations ofterrestrial wildlife still exist, although somefur-bearing species are subject to huntingand trapping, both licit and illicit.

Today the YNAO is numerically theworld’s most productive reindeer herdingregion with 556 000 domestic reindeer,herded by approximately 13 000 nomadicNenets and to a lesser extent Komi andKhanty families. While successor enterprisesof Soviet state farms, sovkhoz dominate theinstitutional landscape in reindeer herding,private clan communities dynamicallydevelop in the North of the okrug and holdmore than half of the reindeer population.At the same time, the gas deposits discoveredhere are the largest worldwide. Large-scaleindustrialisation since the 1970’s has led toan influx of people from the south, whichis why today the indigenous share of theoverall population is less than 7%.


21

Timo Kumpula & al.

This paper focuses on the vastBovanenkovo gas field in the northwesternpart of the Yamal Peninsula (Fig. 1). Thegas field is located in the area to which theYarsalinski reindeer sovkhoz holds theprincipal land title. Bovanenkovo is on themigration path of two major Yarsalinskireindeer brigades. These brigades reachBovanenkovo from south in early-mid Julyon the way to the Kara Sea coast wherereindeer are taken to for insect relief andhigh quality forage. Brigades return throughBovanenkovo in mid-late August when theystart migration towards winter pastures onthe south side of the Ob river. In addition,at least a dozen private reindeer herdingcamps use the larger Bovanenkovo area asreindeer grazing grounds. In comparison tothe brigades, their herds are smaller in size,their migration routes shorter, and their

presence in Bovanenkovo area longer,reaching from early summer until October-November. Very little is known about theexact migration patterns of these privateherders. However, their number has beenincreasing after the fall of the Soviet Union,and their herds as well as their familiesprobably outnumber those of the officiallyregistered herding brigades today.

Materials and methods

Field surveys

Two field expeditions to Bovanenkovo gasfield and to its surroundings were made

Figure 1. Reindeer herding in northern Fennoscandia and Siberia. Bovanenkovo gas field in YamalPeninsula is marked with a black circle (lat. 70° 20’N long. 68°30’E). (Map: Arctic Centre, University ofLapland)


22

during the summers 2004 and 2005. A fewhundred field sites were surveyed for imageclassification. Vegetation types were definedand special attention was placed on the signsof human impacts. Brigades migratingthrough the area were interviewed and alsogas workers from Bovanenkovo wereinterviewed. Valuable information aboutBovanenkovo’s history and exploration weregained from both reindeer herders and gasworkers. Herders from two brigades wereaccompanied during their migration throughthe gas deposit, and explained the migrationhistory, their relations with the gas workers,changes and especially effects of gasexploration on everyday reindeer herdingand reindeer pastures. Printouts of satelliteimages and topographical maps were usedin interviews. Information from herders wasalso obtained during participant observationand in depth unstructured interviews thatprovided deeper insights about the relationsof herders to the land, including its spiritualsignificance and the impact that thepresence of industry has on these relations.

Remote sensing and GIS data andmethods

Satellite images with different spatialresolution were used to compare thevisibility of the different impacts. The veryhigh resolution remote sensing data used inthis study was Quickbird-2 (dated 15.7.2004).The spatial resolution of Quickbird-2multispectral channels is 2.5 meters and onthe panchromatic channel it is 0.61 cm.Quickbird-2 has 4 multispectral channels ofblue, green, red and infrared. Data arestored in 11 byte format. High resolutiondata used were ASTER TERRA with 15

meter resolution (4 channels) (dated21.7.2001), Landsat TM with 30 meterresolution (7 channels) (dated 07.08.1988),and Landsat MSS with 80 meter resolution(4 channels) (dated 28.7.1984) (Table 1).

GIS data with adequate accuracy werenot available. Road network, sand quarriesand drilling sites were digitized usingQuickbird-2 and ASTER TERRA images.Visibility of different impacts on satelliteimages was compared between the imagesusing field sites, ground photographs andphotographs taken from helicopter.Reindeer herders’ migration routes andbrigadiers’ notes were digitized fromsatellite image printouts. The aim was toproduce a map depicting the total area ofdisturbance around the Bovanenkovo.Impact area estimation was produced byusing field sites, photographs taken fromhelicopter and by visual interpretation ofASTER TERRA and Quickbird-2 images.Estimation was carried out by outlining thefurthest visible signs of industrial impactsfrom around Bovanenkovo gas field. Thearea outlined in the ASTER TERRA imageencompasses all visible signs of disturbance,such as roads, off-road vehicle tracks,quarries, etc.

Results

Detecting the impacts

From the remote sensing point of view itis always easier to detect the surfacecharacteristics when trees and their canopyand shadows are absent. The detection ofimpacts from both the satellite data used


23

Timo Kumpula & al.

and the field surveys is presented in Table2. Contamination of the soil because of oiland various chemicals is difficult to detecteven in the field if the contaminated area issmall or if accidents have occurred sometime ago and have since revegetated. Onlya few possible sites were found where therewere some sign of oil spills but which musthave been leaked from a vehicle or an oilbarrel (Bovanenkovo is a gas field). In satelliteimages there were no visible signs of oil orchemical contamination.

The removal of top soil and vegetationtakes place in areas that have been used asbuilding sites, vehicle storage areas, drillingsites activity etc. They usually extend oversome acres or hectares and can be detectedeven from coarse resolution Landsat TMimages. Larger quarries where sand is mined

for road and other building purposes aredetectable also from Landsat MSS images.This is because the spectral reflectance ofbare ground is significantly differentcompared to the reflectance of vegetation.

Garbage is unfortunately difficult todetect. In the field there can be piles ofleftover wood and metal gathered togetherthat is possible to detect from panchromaticQuickbird-2 imagery. Wood remaining onabandoned drilling sites is of high value toreindeer herders who have to cope with ascarcity of wood for cooking and otherpurposes. Herders also make extensiveinnovative use of other leftovers on formerindustrial facilities and by doing socontribute to the recycling of waste to acertain extent (Stammler 2002). When metaland glass waste is sparsely distributed,

Table 1. Spectral characteristics of the satellite data used.

dnablartcepS2-dribkciuQ

)mµ(htgnelevaWARRETRETSA

)mµ(htgnelevaWMTtasdnaL

)mµ(htgnelevaWSSMtasdnaL

)mµ(htgnelevaW

)eulb(1 25.0-54.0 - 25.0-54.0 6.0-5.0

)neerg(2 06.0-25.0 36.0-25.0 06.0-25.0 7.0-6.0

)der(3 96.0-36.0 96.0-36.0 96.0-36.0 8.0-7.0

)RIN(4 98.0-67.0 68.0-67.0 09.0-67.0 1.1-8.0

)RIN(5 - - 57.1-55.1 -

)RIT(6 - - 05.21-04.01 -

)RIN(7 - - 53.2-80.2 -

citamorhcnaP 09.0-54.0 - - -


24

detecting them in field survey can bedifficult. Garbage can partly or completelybe covered with vegetation. This kind ofgarbage is one of the most harmful toreindeer according to herders. Reindeer caninjure their hooves from glass or sharpmetal and hence be exposed to infectionsthat may in the worst scenario lead to death.

Pipelines and power lines are reasonablydetectable in multispectral Quickbird-2imagery. Roads, winter roads and off-roadtracks are just barely visible in Landsat MSSimage. Drill towers and barracks can be

detected with ASTER TERRA imagery.Vehicles can also be seen from multispectralQuickbird images.

Field survey was the most useful way toidentify different impacts. Some objects likerevegetated old tracks, some seldom usedoff-road tracks and areas that have beenrevegetated completely were easier todetect and outline from multispectralQuickbird-2 image. Noticeable is that sometracks were impossible to detect in the fieldbefore we checked the satellite imageprintout.

Table 2. Capacity to detect different impacts of petroleum exploration in Bovanenkovo. Scale is –not vi-sible, x visible with effort, xx quite visible, xxx clearly visible.

otyticapaCtcapmItceted

dleiFyevrus

2-dribkciuQcitamorhcnaP

2-dribkciuQlartcepsitluM

RETSARINVARRET

tasdnaLMT

tasdnaLSSM

,noitanimatnoclioSslacimehc&lio

x - - - - -

liospotfolavomeRnoitategevdna

xxx xxx xxx xx x x

seirrauQ xxx xxx xxx xxx xxx

egabraG

latem- xx - - - - -

ssalg- x-

- - - -

etercnoc- xxx x x - - -

doow- xxx x - - - -

senilepiP xxx xx x - - -

senilrewoP xxx xx x - - -

sdaoR xxx xxx xxx xxx x x

elcihevdaor-ffOskcart

xx xxx xx xx x x

sdaorretniW xx xx xx xx x -

srewotllirD xxx xxx xx x - -

skcarraB xxx xxx xx x - -

selciheV/skcurT xxx xx x - - -


25

Timo Kumpula & al.

Scales of investigations

When investigating environmental impacts,the data resolution has to be suitable forresearch purposes. For example pipelinescannot be detected from Landsat TMimages accurately. Size of the research areais another factor determining the data to beused either in terms of data size (megabytes)or data costs. In the case of Bovanenkovo,where the total area is 35 * 10 km, ASTERTERRA is the most cost efficient datasource to study impacts of the whole gasfield. But if more detailed study is planned,Quicbird-2 image is a very good choice.Quickbird-2 multispectral image enhancedwith panchromatic image provide the bestdata to detect impacts around drilling sites.Most impacts are detectable in this imagecombination (Fig. 2).

ASTER TERRA images are worthy ofattention from among the larger scalesources of satellite data. Landsat TMcovers a larger area and has better spectralresolution with 7 channels, but ASTERTERRA’s 15 meter resolution and lowerprice increases its overall utility to detectenvironmental impacts.

Area of disturbance

Total area of disturbance was calculated byoutlining inside Bovanenkovo the areas thathad signs of human disturbances visible inASTER TERRA and Quickbird-2 images(Fig. 3). The area of disturbance covers 317km2. Inside this buffer there are visiblechanges compared to the original vegetationcover. Inside this area there is also a significant

Figure 2. Scales of investigations and relevant remote sensing data sources (Timo Kumpula & BenjaminBurkhard).


26

amount of relatively undisturbed vegetation.Permanently lost pasture-land includes roadsand infrastructure areas. The active roadnetwork totals 79 km and it covers 143 ha,by calculating an 18 m wide buffer aroundroad network. Although roads are onlyapproximately 18 m wide, the actual affectedzone around them is much broader due tothe cumulative impacts of blowing sandand dust and altered hydrology. Significantchanges in soil moisture, pH and plantchemistry are detectable at least 35 m fromthe edges of roads within four years of theirconstruction (Forbes 1995). Infrastructure(drill sites, barracks, storage areas) covers203 ha and quarries 373 ha. However, weare aware that besides these detectableimpacts on satellite images, reindeer herdersare heavily impacted by the social andeconomic consequences of industrialinfrastructure: for example, a road opens

up larger areas of the tundra for openaccess by industrial workers. Besides forworking purposes, they may use roads forgoing on hunting, fishing (poaching),recreational trips or trading. Being untrainedfor the interaction with herders and herds,they might not behave in an appropriateway in the vicinity of herders and herds.Therefore, roads are significant for herdersbeyond the immediate corridor of impactedpasture.

Herders’ and workers’ interpretation ofsatellite images

Interviewing reindeer herders was essentialto understanding the impacts of petroleumindustry on reindeer and their herders.Information on the history of exploration

Figure 3. Area of disturbance around Bovanenkovo gas field.


27

Timo Kumpula & al.

which has taken place in the area would havebeen extremely difficult or totallyimpossible to gain from other sources. Manyherders were able to interpret satelliteimage printouts extremely efficiently (Fig.4). False colour composites did not causeproblems to identify lakes, rivers, campsitesand migration routes with their exactlocations. Herders through their perceptionof the migration route and the industrialarea added an important level of historicaldepth and detail to our understanding ofrecent changes in the area. Gas workers alsoprovided a lot of information about thearea and history from their point of view.They also provided information concerningthe future of the petroleum exploration inthe area. Interviews and participantobservation with both groups of workersin the tundra impressively revealed thevarious mental maps of people withdifferent interests in the area. The mentalmap of most herders consists of a complexset of migration routes, pastures, fishinggrounds, camp sites presenting a detailedimage of the on-surface resources. The gasworkers’ mental map in the Bovanenkovoarea was also very detailed and in many casesreached back decades, consisting of acomplex set of drilling sites that provideaccess to different layers of the gas underthe permafrost, and of pipelines, roads, andcamps dispersed across the whole territoryof the gas deposit. Gas workers also haddeveloped throughout the decades anintimate knowledge of the rivers and lakes,some of which they have been using forfishing for a long time, and some of whichare now being heavily impacted by newroad constructions leading right through thewater bodies.

Figure 4. Brigadier Vasili Serotetto explainingmigration routes, campsites and pastures withfalse-colour composite of ASTER TERRA imagery.Photo: Timo Kumpula 14.7.2005.

The technology of building bridges byblocking off small rivers or lakes completelyfor the duration of the road works has beenheavily criticised by herders. As a result,larger areas around these blocked rivers canbecome flooded and turn into extremewetlands that are not suitable as reindeerpasture or migration routes anymore.These flooded areas as well as drained lakescan also be detected on satellite pictures andherders added historical depth and detaileddescriptions of events for these areas.


28

Discussion

The preliminary results in detecting theimpacts of oil petroleum exploration onreindeer herding were promising. It seemsthat combining different scale remotesensing data results in impacts detection canbe achieved. Usually drillings sites arespatially quite small and like Bovanenkovothe whole gas field is 35 * 15 km. This allowspinpointing the special area of interest andto acquire expensive very high resolution datalike Quickbird-2 or IKONOS-2. Thendetailed research on impacts is possible toconduct.

During the field survey the limitations ofhuman eye were once more clearlydemonstrated. We noticed that in the fieldit was difficult to notice revegetated vehicletracks and the edges between once disturbedand untouched vegetation. In satellite imageprintouts different intensities of infra-redsignatures were very clear. Also detectingold vehicle tracks from ground level wasmuch more difficult than from helicopterand Quickbird-2 images.

Local and indigenous peoples’ assessmentwas crucial in understanding currentdevelopment and impacts of petroleumexploration. In general, they add a differentlevel of interpretation to changes on thetundra by ranking events, development andchanges and evaluating them in terms oftheir positive or negative impact on theirlivelihoods. Thus, satellite images andbiological ground truthing provide animportant factual basis and informationabout their significance for the lives ofhumans and animals on the tundra. Herdersare able to provide qualitative data (testimonyfrom active and retired herders) on the degree

of environmental and socio-economicchange during past decades. They provideinformation on the current variation in thevegetation and reindeer habitat types andhow they are ranked according to quality.They can also estimate what proportion ofthe habitat has been lost because of thepetroleum industry and how the herdersand reindeer have adapted to this. Herdersalso have a good idea where areas ofpotential or likely future degradation mayoccur (e.g. desertification from heavygrazing, melting of frozen ground). Herders’expectations or scenarios for the future underpetroleum development are importantfactors that should be considered in researchquestion development. Nenets developedstrategies for mitigation of predicted negativechanges and scenarios of future reindeermanagement in conjunction with petroleumdevelopment.

What are more difficult or impossible todetect on satellite images are the potentialand actual positive social and economicimpacts of industrial development onreindeer herders. During fieldwork,participant observation with them revealedthat in principle the presence of industrialworkers and their facilities is welcomed byherders of the area, and their perception ofthe negative effects is balanced by thebenefits, for example of the workers’settlements as a source of staple food toresupply, the use of roads for the supply ofherders, the settlements as a market forreindeer meat, the helicopters of the gascompany as means of transport and asmeans to collect velvet reindeer antlers tobe sold on far eastern markets for cash(Stammler 2004). In the case of the privateherding camps not belonging to officiallyregistered reindeer brigades, the gas company


29

Timo Kumpula & al.

has become the most important tradingpartner.

The further processing of satellite datafocuses on the extent and rate of tundraecosystem change during the past 20 – 30+years. Landsat MSS and TM are the maindata sources when studying the situation inthe 1970’s and 1980’s but also CORONAspy satellite images can be used in changedetection. The aim is to estimate how largea fraction of the habitat has been affectedwithin the current range of reindeer habitats.GIS and risk assessment of potential futuredegradation in expanding drilling sites willbe conducted.

Together the applied results from theseassessments will provide a potentiallypowerful tool for regional planning andpolicy development. The results will be ofinterest to scientists in several disciplines (e.g.anthropology, geography, biology/ecology,political science, international relations,conservation biology), as well as to indigenousgroups elsewhere in other arctic and subarcticregions.

Acknowledgements

We would like to thank the Finnish Academyfor financing the project Environmetal andSocial Impacts of Industrial Developmentin Northern Russia (ENSINOR) 2004-2007.

References

Allard, A. (2003). Detection of vegetationdegradation on Swedish mountainous heathsat an early stage by image interpretation. Ambiovol. 32:8, 510-519.

Benelli, G. & A. Garzelli (1999). Oil-spills detectionin SAR images by fractal dimension estimation,Proc. IGARSS’99 vol. 1, 218–220.

Espedal, H.A. & O.M. Johannessen (2000).Detection of oil spills near offshore installationsusing synthetic aperture radar (SAR).International Journal of Remote Sensing vol.21: 11, 2141–2144.

Forbes, B.C. (1995). Tundra disturbance studiesIII. Short-term effects of aeolian sand and dust,Yamal Region, northwest Siberia, Russia.Environmental Conservation vol. 22, 335-344.

Forbes, B.C. (1999). Reindeer herding andpetroleum development on Poluostrov Yamal:sustainable or mutually incompatible uses?Polar Record vol. 35, 317-322.

Forbes, B.C. (2004). Impacts of energydevelopment in polar regions. In C.J. Cleveland(ed.): Encyclopedia of Energy , 93-105.Academic Press, San Diego.

Ingold, T. (2005). A manifesto for the anthropologyof the North. In Sudkamp A. (ed.): Connections:local and global aspects of Arctic socialsystems, 61-71. University of Alaska,Fairbanks.

Jernsletten, J.-L. & K. Klokov (2002). Sustainablereindeer husbandry. Arctic Council/Centre forSaami Studies, Tromsø.

Jones, B.A. (2001). Comparison of visualobservations of surface oil with syntheticaperture radar imagery of the Sea Empress oilspill. International Journal of Remote Sensingvol. 22: 9, 1619–1638.

Klein, D.R. (2000). Arctic grazing systems andindustrial development: can we minimizeconflicts? Polar Research vol. 19, 91-98.

Krupnik, I. (2000). Reindeer pastoralism in modernSiberia: research and survival in the time ofcrash. Polar Research vol. 19, 49-56.


30

Kumpula, T. (2006). Very High Resolution RemoteSensing Data in Reindeer Pasture Inventory inNorthern Fennoscandia. In Forbes, B.C., M.Bölter, L. Müller-Wille, J. Hukkinen, F. Müller, N.Gunslay & Y. Konstantinov (eds.): ReindeerManagement in northernmost Europe, 167-185.Springer, Berlin.

Meschtyb, N.A., B.C. Forbes & P. Kankaanpää(2005). Social Impact Assessment alongRussia’s Northern Sea Route: petroleumtransport and the Arctic Operational Platform(ARCOP). Arctic InfoNorth 58, 322-327.

Mosnews (2006). Russia Overtakes Saudi Arabiaas World’s Leading Oil Producer. 11.09.2006.http://www.mosnews.com/money/2006/08/23/russiaoil.shtml.

Pika, A. & D. Bogoyavlensky (1995). YamalPeninsula: oil and gas development andproblems of demography and health amongindigenous populations. Arctic Anthropologyvol. 32, 61-74.

Rees, W.G., M. Williams & P. Vitebsky (2003).Mapping land cover change in a reindeerherding area of the Russian Arctic usingLandsat TM and ETM+ imagery and indigenousknowledge. Remote Sensing of Environmentvol. 85: 4, 441-452.

Solberg, A.H.S., S.T. Dokken & R. Solberg (2003).Automatic detection of oil spills in Envisat,Radarsat and ERS SAR images. Proc.IGARSS’03 vol. 4, 2747–2749.

Stammler, F. (2002). Success at the edge of theland: Present and past challenges for reindeerherders of the West-Siberian Yamal-Nenetsautonomous Okrug. Nomadic Peoples NS vol.6, 51-71.

Stammler, F. (2004). The Commoditisation ofReindeer Herding in Post Soviet Russia:Herders, Antlers and Traders in Yamal. In LederS. & B. Streck (eds.): Segmentation undKomplementaritaet. Organisatorische,oekonomische und kulturelle Aspekte derInteraktion von Nomaden und Sesshaften.Mitteilungen des SFB “Differenz undIntegration”. Orientwissenschaftliche Hefte vol.14, 105-122.

Stammler, F. (2005). Reindeer Nomads Meet theMarket: Culture, Property and Globalisation atthe End of the Land. Halle Studies in theAnthropology of Eurasia vol. 6. Lit publishers,Münster.

Stammler, F. & E. Wilson (2006). Dialogue fordevelopment: an exploration of relationsbetween oil and gas companies, communitiesand the state. In Wilson E. & F. Stammler (eds.):Special issue on the oil and gas industry localcommunities and the state, 1-42. Sibirica vol.5: 2.

Stow, D.A., A. Hope, D. McGuire, D. Verbyla, J.Gamon, F. Huemmrich, S. Houston, C. Racine,M. Sturm, K. Tape, L. Hinzman, K. Yoshikawa,C. Tweedie, B. Noyle, C. Silapaswan, D.Douglas, B. Griffith, G. Jia, H. Epstein, D. Walker,S. Daeschner, A. Petersen, L.M Zhou & R.Myneni (2004). Remote sensing of vegetationand land-cover change in Arctic TundraEcosystems. Remote Sensing of Environmentvol. 89, 281-308.

Toutoubalina, O.V. & W.G. Rees (1999). Remotesensing in detection of industrial impact aroundNoril’sk, northern Siberia: preliminary results.International Journal of Remote Sensing vol.20: 15-16, 2979-2990.

Tømmervik, H., K.A. Høgda & I. Solheim (2003).Monitoring vegetation changes in Pasvik(Norway) and Pechenga in Kola peninsula(Russia) using multitemporal landsat MSS/TMdata. Remote sensing of Environment vol. 85,370–388.

Virtanen, T., K. Mikkola, E. Patova & A. Nikula(2002). Satellite image analysis of humancaused changes in the tundra vegetationaround the city of Vorkuta, north-EuropeanRussia. Environmental Pollution vol. 120: 3,647-658.

Zen’ko, M. A. (2004). Contemporary Yamal:ethnoecological and ethnosocial problems.Anthropology & Archaeology of Eurasia vol. 42,7-63.

combining data from satellite images and reindeer herders in … · 2011. 4. 29. · and practical...

Documents