research article a webgis-based information system for...

Hindawi Publishing CorporationAdvances in MeteorologyVolume 2013 Article ID 769270 9 pageshttpdxdoiorg1011552013769270

Research ArticleA WebGIS-Based Information System for Monitoring andWarning of Geological Disasters for Lanzhou City China

Fang Miao and Qi Yuan

Cold and Arid Regions Environmental and Engineering Research Institute Chinese Academy of Sciences (CAS)Lanzhou 730000 China

Correspondence should be addressed to Qi Yuan qiyuanlzbaccn

Received 5 September 2013 Revised 6 November 2013 Accepted 27 November 2013

Academic Editor Chung-Ru Ho

Copyright copy 2013 F Miao and Q YuanThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Monitoring and warning of geological disasters accurately and in a timely fashion would dramatically mitigate casualties andeconomic losses This paper takes Lanzhou city as an example and designs a Web-based system namely the information systemfor geological disaster monitoring and warning (ISGDMW) Presented are its framework key developing technologies databaseand working flow The information system adopts a BrowserServer (BS) structure and has three-tier architecture combining in-situ monitoring instruments the wireless sensor network WebGIS techniques and the grey system theory The framework of theISGDMW can be divided into three categories (1) in-situ monitoring system it aims to monitor geological disaster sites and getstate information of geological disaster sites (2) database manage in-situ monitoring data antecedent field investigating data andbasic data (3) analyzing and warning system analyze in-situ monitoring data understand the deformation trend of the potentialgeological disaster and release disaster warning information to the publicThe ISGDMW allow the processes of geological disastermonitoring in-situ monitoring data analysis geological disaster warning to be implemented in an efficient and quick way and canprovide scientific suggestions to commanders for quick response to the possibility of geological disaster

1 Introduction

To mitigate geological disaster we should depend on bothreal-time in situ data and quick response to the possibilityof geological disaster WebGIS is the integrated product ofgeographic information system (GIS) and internet technolo-gies in WebGIS the internet technologies are connectedwith GIS in order to take advantage of their special char-acteristics such as easy usability use of the GIS data suchas input adjustment manipulation analysis and output ofgeographical information and to bring out related service onthe internetWhereas previous standalone GIS had restrictedapplication capability on the network the WebGIS makes itpossible to retrieve and analyze spatial data through theweb The internet also provides a medium for processinggeorelated information with no location restrictions [1] Inaddition WebGIS promotes the sharing and synthesis ofmultisource data and enables widespread sharing of spatialdata and geosciences models [2] Therefore WebGIS offersa powerful and advanced approach to prevent and mitigate

geological disaster and it has played a significant role interms of transmitting catastrophe data analyzing the disastercondition and releasing disaster information [3ndash7] In situmonitoring data can be used to grasp the deformation trendof the geological disaster therefore in situ monitoring instr-uments (eg inclinometers rain gauges piezometers andextensometers) must be playing an important role in theprocess of geological disaster mitigation [7ndash10]

In this paper we comprehensively utilize the advanta-ges of in situ monitoring instruments the wireless sensornetwork and WebGIS techniques in terms of in situ moni-toring transmitting data disaster analysis and data man-agement to design a Web-based system namely the infor-mation system for geological disaster monitoring and warn-ing (ISGDMW) Deformation trends of the landslide anddebris flow are analyzed automatically with the grey sys-tem method and threshold rainfall respectively based onISGDMW ISGDMW is a novel platform and is designed toimprove practical efficiency of geological disaster mitigationin Lanzhou city

2 Advances in Meteorology

Map of China

Islands of South China Sea

Map of Gansu

Debris flowLandslideYellow river

FaultMain roadCity boundary

ElevationHigh 3670 m

Low 1440 m

Qinghai-Tibetan Plateau

Loess Plateau

N0 10 20 40

(km)

103∘09984000998400998400E 103

∘30

9984000998400998400E 104

∘09984000998400998400E 104

∘30

9984000998400998400E

103∘09984000998400998400E 103

∘30

9984000998400998400E 104

∘09984000998400998400E 104

∘30

9984000998400998400E

37∘09984000998400998400N

36∘309984000998400998400N

36∘09984000998400998400N

37∘09984000998400998400N

36∘309984000998400998400N

36∘09984000998400998400N

Province

Figure 1 The study area landslide sites and debris flow sites with high risk were investigated and supplied by the Lanzhou Bureau of Landand Resources

2 Study Area

Lanzhou city is located in the transitional zone between theQinghai-Tibet Plateau and the Loess Plateau (Figure 1) Inthis area geological conditions are very complex and somedeep and large gullies exist in most parts of the territoryThese geological and geographic backgrounds lead to Lan-zhou being prone to landslides and debris flow disa-sters The latest statistics [11] show that the existing landsl-ide sites (including unstable slopes) number 417 in Lanz-hou In addition in Lan zhou the rainfall from Aprilto September accounts for about 868 of the annual pre-cipitation [11] and always occurs in the form of heavy rainand storms This pattern of rainfall is consistent with thefact that landslide disasters and debris flow always occurduring the months of April to September in Lanzhou Inthe last 5 years (from 2008 to 2012) 16 serious landslidedisasters broke out in this city each landslide disaster causedeconomic losses of more than 01 million CNY 14 among the16 serious landslide disasters were triggered by heavy rainfalland the other 2 landslide disasters were triggered by snowand ice melt detailed presentation is shown in Table 1 At1800 Beijing standard time May 16 2009 the most seriouslandslide disaster occurred at Jiuzhou district in LanzhouThis landslide disaster caused the death of 7 people and the 1person injury of and the direct economic loss of 206 millionCNY was recorded The Jiuzhou landslide belongs to loesslandslide categoryThe length and width of Jiuzhou landslidewere 160 and 40 meters respectively and covered an area ofabout 7500m2 with an average depth of about 40metersThetotal volume was about 62 times 104m3 Figure 2 demonstratesthe front view of the Jiuzhou landslide

3 Landslide and Debris FlowForecasting Method

31 Grey System Forecasting of Landslide Deformation Thegrey system theory was initiated by Deng [12] The conceptof the grey system in its theory and successful applicationis now well known in China It is able to (1) analyze theindeterminate and incomplete data to establish the systematicrelations [13] and (2) forecast time series accurately andthis has been quite a popular subject for researchers bothin the past and at present [14] It assumes that the internalstructure parameters and characteristics of the observedsystem are unknown The system state can be predicted by adifferential equation from the recent historicalmeasurements[13] Although the historical measurements are too complexor chaotic they always contain some governing laws [14]The grey prediction has been widely used in applicationsof geography [15] agriculture [16] runoff prediction [17]displacement prediction of landslide [18] prediction of slopestability [19] power demand [20] stock market [21] and soforthThis theory is also coupled with theWebGIS to forecasttimely deformation in the field of engineering geology suchas the deformation of landslides and host rock in the cavity[7]

The model GM (1 1) one of the grey models is themost widely used in the literature pronounced as ldquogreymodel first order one variablerdquo This model is a time seriesforecasting model The differential equations of the GM (11) model have time varying coefficients In other wordsthe model is renewed as the new data become available tothe prediction model The GM (1 1) model can only beused in positive data sequences [22] In this paper since all

Advances in Meteorology 3

Table 1 Serious landslide disaster records in Lanzhou from 2008 to 2012 It was recorded and supplied by the Lanzhou Bureau of Land andResources

Occurrence date Site Trigger factor Casualties Economic losses (01 million CNY)March 28 2008 Chengguan district Snow and ice melt 1 200July 18 2008 Gaolan county Rainfall 1 100September 19 2008 Gaolan county Rainfall 50May 16 2009 Chengguan district Rainfall 8 20600September 14 2009 Chengguan district Rainfall 3 800September 15 2009 Qilihe district Rainfall 300October 22 2009 Qilihe district Rainfall 1300August 16 2010 Honggu district Rainfall 10October 19 2010 Chengguan district Rainfall 10April 26 2011 Xigu district Rainfall 100August 29 2011 Yuzhong county Rainfall 16October 8 2011 Xigu district Rainfall 86March 17 2012 Chengguan district Snow and ice melt 600April 20 2012 Yuzhong county Rainfall 30May 21 2012 Chengguan district Rainfall 50May 21 2012 Chengguan district Rainfall 100lowastSource from Lanzhou Bureau of Land and Resources

(a) (b)

Figure 2 Jiuzhou landslide outburst at 1800 Beijing standard time May 16 2009 The left image is the remote viewing of Jiuzhou landslidethe right image is the close viewing of Jiuzhou landslide

the primitive data points are positive grey models can beused to forecast the trend of deformation of landslide in thispaper

The basic procedure for grey prediction is listed as follows[23]

Step 1 Construct a data series that contains the recentlymea-sured displacement of a landslide

119911(0)

= 119911(0)

(1) 119911(0)

(2) 119911(0)

(119899)

= 119911(0)

(119896) 119896 = 1 2 119899

(1)

where 119911(0)

(119896) is the measurement from sensory informationat time 119896 and 119899 is the length of the data series

Step 2 Form a new data series 119911(1) by an accumulated gene-

rating operation (AGO)

119911(1)

= 119911(1)

(1) 119911(1)

(2) 119911(1)

(119899)

= 119911(1)

(119896) 119896 = 1 2 119899

(2)

where

119911(1)

(119896) =

119896

sum

119894=1

119911(0)

(119894) 119896 = 1 2 119899 (3)

Step 3 Form the grey differential equation

119889119911(1)

119889119905

+ 119886119911(1)

= 119887 (4)


with initial condition 119911(1)

(1) = 119911(0)

(1) The coefficients 119886 and119887 can be obtained by using the least squaremethod as shownin

119886 = [119886

119887] = (119861

119879119861)

minus1

119861119879119884 (5)

where

119861 =

[[[

[

minus119885(1)

(2) 1

minus119885(1)

(3) 1

sdot sdot sdot sdot sdot sdot

minus119885(1)

(119899) 1

]]]

]

119884 =

[[[

[

119911(0)

(2)

119911(0)

(3)

sdot sdot sdot

119911(0)

(119899)

]]]

]

(6)

119885(1)

(119896) = 120572119911(1)

(119896) + (1 minus 120572)119911(1)

(119896 minus 1) 119896 = 2 3 119899 and 120572 isthe weighting factor according to relevant literature [13] inthis paper we specify 120572 is a constant with 05

Step 4 (obtain the prediction value) Once 119886 and 119887 in (4) areobtained the grey differential equation can be used to predictthe value of state 119911 at time instant 119896 + 1

The AGO grey prediction model can be obtained

(1)

(119896 + 1) = [119911(0)

(1) minus

119887

119886

] 119890minus119886119896

+

119887

119886

119896 = 0 1 (7)

Then the prediction value of the state can be calculated byan inverse accumulated generating operation (IAGO)

(0)

(119896 + 1) = (1)

(119896 + 1) minus (1)

(119896)

= (1 minus 119890minus119886

) [119911(0)

(1) minus

119887

119886

] 119890minus119886119896

(8)

For instance the following figure (Figure 3) that includes13 ensembles the former 11 ensembles which represent themeasured displacement of a landslide and the latter 2 ense-mbles that represent the forecasted displacement of the land-slide based on both the basic procedure of grey predictiontheory and the measured displacement of the landslide

32 Critical Rainfall of Debris Flow Forecasting Real-timeassessment of debris flow disaster is fundamental for buildingwarning systems that can mitigate its risk A convenientmethod to assess the possible occurrence of a debris flowis the comparison of measured and forecasted rainfall withrainfall threshold curves (RTC) [24]Therefore how to definethe RTC is a key issue in order to prepare efficient forecastingin amountainous region (eg Lanzhou) that is prone to rain-triggered debris flow

Rainfall especially heavy rainfall is the most criticalnatural triggering factor in Lanzhou Rainfall intensity andduration of storms have been shown to influence the trigger-ing of debris flows The relationship between intense rainfalland debris flow initiation has been widely analyzed anddocumented in the literature in a number of different settingsand environments throughout the world [25]

To define triggering thresholds Bacchini and Zannoni[25] compared rainfall data to the occurrence of debris flowsto examine the relations between debris flow initiation and

0102030405060708090

100

Disp

lace

men

t (m

m)

Augu

st 10

201

0

Augu

st 20

201

0

Augu

st 30

201

0

Sept

embe

r 10

201

0

Sept

embe

r 20

201

0

Sept

embe

r 30

201

0

Oct

ober

10

201

0

Oct

ober

20

201

0

Oct

ober

30

201

0

Nov

embe

r 10

201

0

Nov

embe

r 20

201

0Fo

reca

sted

Nov

embe

r 30

201

0Fo

reca

sted

Dec

embe

r 10

201

0

Figure 3 Data series of measured and forecasted displacementsof a landslide assuming that the current date is November 202010 from August 10 2010 to November 20 2010 the landslidedisplacementmetermeasured a data series that contains 11measureddisplacements of a landslide based on the formulas (1)ndash(8) and11 measured displacements It can calculate and forecast landslidedisplacement value in the next 10 days (November 20 2010ndashNovember 30 2010 and November 30 2010ndashDecember 10 2010)Black solid circles denote the measured displacement of a landslideblack solid triangle points denote the forecasted displacement of thelandslide

0

5

10

15

20

25

30

16 05 1 3 6 24

Accu

mul

ated

rain

fall

(mm

)

Rainfall duration (h)

Figure 4 Rainfall threshold curve

rainfall in the area of Cancia (Dolomites Northeastern Italy)Tan and Duan [26] had studied the relation between debrisflow initiation and minimum rainfall in China and prelim-inarily defined the RTC but the result was not as accurateas possible due to the fact that it is a large-scale result Wuet al [27] considered that local condition especially naturalcondition (eg precipitation topography and geology) is akey factor to assess this issue simultaneously other relevantstudies [26] should not be neglected in the research Underthis methodology Wu et al [27] had identified the criticalrainfall of debris flow initiation in LanzhouThe new advancepublished by Wu et al [27] is a local-scale result and asaccurate as possible thus it is suitable to forecast debrisflow in Lanzhou and be coupled with ISGDMW detailedinformation about the new progress is presented in Table 2and Figure 4

According to Table 2 in Lanzhou city the critical rainfallin 10 minutes 30 minutes 1 hour 3 hours 6 hours and 24hours is 5mm 7mm 8mm 18mm 24mm and 25mmrespectively RTC represents the relationship between rainfall


Table 2 Critical rainfall to trigger debris flow in Lanzhou

Rainfall duration (h) Accumulated rainfall (mm)16 505 71 83 186 2424 25

duration and critical rainfall From Table 2 we plotted theRTC which is especially suitable for Lanzhou (Figure 4)When it rains the algorithm of comparing the latest accu-mulated rainfall value measured by rain gauge and therainfall threshold curve is implemented automatically in theanalyzing module of ISGDMW If the latest accumulatedrainfall value is located above the curves this debris flow siteis considered to be dangerous It is important to emphasizethat Wu et al [27] had only studied the relationship betweenrainfall duration and critical rainfall within 24 hours due tothe shortage of data So in case the rainfall duration is over24 hours as long as the accumulated rainfall is greater than25mm the corresponding debris flow site is still consideredto be dangerous the same conclusion is also applicable tosituation of rainfall duration being less than 10 minutes butthe accumulated rainfall being more than 5mm

4 System Design

41 Framework of ISGDMW The framework of ISGDMW isshown in Figure 5 and can be divided into three parts

(1) Part 1 In situ monitoring system it mainly includeslandslide monitoring instruments debris flow mon-itoring instruments and wireless sensor networkThese monitoring instruments were installed into thegeological disaster sites and get up-to-date informa-tion of geological disaster sites In situ monitoringdata should be transmitted to the data center anduploaded to the system database through the wirelesssensor network

(2) Part 2 Database it is used for managing and integrat-ing the spatial and nonspatial data related to geolog-ical disasters Those data include in situ monitoringinformation antecedent field investigating data andbasic data In the database the SQL Server2008 data-base software is often employed as a database plat-form and the ArcSDE middleware which is devel-oped by ESRI Company is chosen as the spacedatabase engine

(3) Part 3 Analyzing and warning system it consists of 4modules basic module analyzing module releasingmodule and monitoring module each module has adifferent function (please see detailed information inFigure 5) The core of analyzing and warning systemis the analyzingmodule the theoretical bases-the greysystem and rainfall threshold were coupled into theanalyzing module so it is the connection between

the theoretical bases and ISGDMW Analyzing andwarning system mainly aims to analyze in situ mon-itoring data and the deformation of the potentialgeological disaster sites Based on the grey system the-ory and measured displacement of landslide defor-mation trend of landslide is analyzed automaticallyby ISGDMW while the current state of debris flowis analyzed through comparing accumulated rainfallwith the rainfall threshold Finally analyzed results(namely output data from ISGDMW) and warninginformation will be released to the public through theinternet E-mail and message based on the releasingmodule of ISGDMW

42 Database of ISGDMW Database of the ISGDMW is thefundamental component On the one hand it assists inmana-gement of the data related to geological disaster effectivelyand on the other hand it provides data support for analyzinggeological disaster conditionsThe system database is dividedinto two categories namely spatial data and nonspatial dataand can be demonstrated in Figure 6 particularly

(i) Spatial Data

(1) Vector layers mainly include administrativemaps land-useland-cover change maps(LUCC) road maps soil maps vegetation typemaps geology maps fault and seismic beltmaps and river maps

(2) Raster layers mainly include digital elevationmodel (DEM) slope gradient maps and vege-tation cover maps

(3) Theme layers mainly include the field surveyingmap of potential geological disaster sites whichare generated by antecedent field surveyingtasks implemented by the Lanzhou Bureau ofLand and Resources In some of those potentialgeological disaster sitesmonitoring instrumentswill be installed in situ

(ii) Nonspatial Data

(1) Monitoring data it mainly includes rainfall datawhich is gathered from rain gauge and displa-cement data of landslide which is measuredby a special instrument-displacement meter forlandslide surface monitorThose data are impo-rted into the database through a wireless sensornetwork

(2) Auxiliary data it involves gross domestic prod-uct (GDP) and the population of every village orcommunity This data is the official reference togeological disaster preventing and control

43Working Flow of ISGDMW According to the sequence ofdata acquirement data analysis and the releasing of warninginformation the process of geological disaster mitigationand prevention based on the platform of ISGDMW can bedesigned in three stages and explained as follows (as shownin Figure 7)


(1) In-situ monitoring system

(2) Database

The information system for geological disaster monitoring

and warning(ISGDMW)

Spatial database

Nonspatialdatabase

Basic moduleMain functions browse basic map geological

disaster sites and disaster information

Monitor moduleMain functions browse monitor instruments

state and monitoring data

Supply data

Upload

Upload

Monitoring instruments

Control module

Solar cell

Wireless sensor network

Analyzing moduleMain functions monitoring data analysis andanalyzing geological disaster site deformation

Releasing moduleMain functions release geological disasterinformation (text and map) to the public

(3) Analyzing and warning system

Potentially geological

disaster sites

Antecedent field investigating

Monitored information Displacement Accumulated rainfall

Geological disaster sites

Monitor

Figure 5 Framework of ISGDMW

Database

Nonspatialdatabase

Vector layer Monitoring data

Spatial database

Auxiliary dataTheme layerRaster layer

∙ Administrative map∙ LUCC∙ Soil map∙ Vegetation∙ Road

∙ DEM∙ Vegetation cover∙ Slope gradient∙ Slope undulation

∙ Geology∙ Fault∙ Seismic belt∙ River

∙ Precipitation∙ Displacement∙ Deformation

∙ GDP∙ Population

∙ Landslide sites∙ Debris flow sites∙ Others

Figure 6 Architecture of database of ISDGMW

(i) Stage One In Situ Monitor and Upload Data Landslidemonitoring instruments automatically measure the displace-ment of landslides andwirelessly send the displacement valueto the nonspatial database every 10 days debris flow moni-toring instruments automatically collect accumulated rainfallwhen it rains and then wirelessly send the accumulatedrainfall value to the nonspatial database every 5 minutes

(ii) Stage Two Analyze In Situ Data The professional man-agers immediately start the analyzing module of SGDMWwhen the latest data from in situ monitoring instrumentsis inputed into the nonspatial database Deformation trendsof landslide are calculated by the grey system methodwhich had been coupled with SGDMW and calculatedresult is the displacement of landslide in the next 10 days


Landslide site Debris flow site

Monitoring data

Database ISGDMW

Results geological disaster state and deformation tendency

Upload

OperationManager

Are there one or some geological disaster sites with high probability

to outburst

Everything remains unchanged

No

Yes

Release disaster information to the public next to thedisaster site(s) Meanwhile start the prevention plan

of geological disaster mitigation

Stag

e one

mon

itor

Stag

e tw

o an

alys

isSt

age t

hree

war

ning

Landslide monitoring instrument

Debris flow monitoring instrument

Description it measures and transmitsautomatically displacement of a landslide to the database every 10 days

Displacement Accumulated rainfall

Sensor landslide displacement meter

Sensor rain gauge

Description it measure and transmit automatically accumulated rainfall to the database every 5 minutes

Figure 7 Working flow of geological disaster monitoring and warning based on ISGDMW

Measured displacement of landslide

Accumulated rainfall

Grey system theory

Rainfall threshold

Forecasted displacement of landslide in the next 10 days

Current state of debris flow site namely dangerous or not

Input data ISGDMW Output data

Figure 8 Data flow of stage two in the working flow of geological disaster monitoring and warning based on ISGDMW

The current state of debris flow is analyzed by comparing theaccumulated rainfall with the threshold rainfall of debris flowoccurrence in Lanzhou if the accumulated rainfall is morethan the threshold rainfall of debris flow occurrence thenthis debris flow site is considered as dangerous That is tosay in this stage the in situ monitoring instrument inputsmeasured displacement of landslide into the database andthen the analyzing module of SGDMW automatically readsthe measured displacement and forecasts the displacement oflandslide in the next 10 days The same process also occurs in

analyzing the current state of debris flowThe data flow of thisstage is shown in Figure 8

(iii) Stage Three Releasing Warning Information For land-slides and debris flows with sharp deformation or dangerousstate the professional manager must send warning messagesto the governor who governs the region which is impactedby the geological disaster site(s) and release warning infor-mation on the internet through the analyzing module ofSGDMW


When the governor receives a warning message he orshe must immediately alarm the public about the disasterwarningmessage bymeans of oral announcement broadcastmobile phone loudspeaker sound the drum (or bell) send-ing out messengers and so forth Meanwhile the governormust start prevention plans for the geological disaster andevacuate people from dangerous sites to safety shelters

44 Developed Technology The ISGDMW adopts a browserserver structure based on a web service and can be dividedinto three tiers namely Data tier Service tier and Appli-cation tier The users in Application tier are acting asterminals via the internet Ordinary users such as the publiccould simply use internet browsers (IE or Firefox) to accessthe released information which the server provides Otherprofessional users such as professional managers could usemore powerful desktop tools to access the server and performsophisticated tasks [10] In the Service tier ArcGIS Server93software which is one of the server GIS products from ESRI(Environmental Systems Research Institute Inc) was chosenas the basic platform for the server application which canbe used to introduce advanced GIS function to the internetenvironment and to publish information based on GIS Inthe Data tier SQL Server2008 software and ArcSDE93 soft-ware are used for managing and integrating the spatial andnonspatial data The Dell server and Windows Server2008operating systemwere used as the application environment ofthe system and theASPnet technologyMicrosoftVS2008netdeveloping environment C programming language andDreamweaver software were chosen as the implementingmeans of the ISGDMW

5 ConclusionsIn general geological disasters in the mountainous area arefrequent and complex in China and in situ monitoring and aquick response are the key methods for mitigating geologicaldisasters in those areas In this paper a WebGIS-basedplatform that is ISGDMW has been designed to enableeffective integration of in situ monitoring data managementgeological disaster analysis sending warning messages andenabling a prompt response The ISGDMW had been imple-mented and tentatively run during the past few months butit still has a little bug in the codes and is kept in checkingWe need to stress that our design scheme of the system isvaluable for others because the system has three significantfeatures including simplicity automation and user friend-liness However since the geological disaster is paroxysmaland complicated ISGDMWmust be further enhanced in twoaspects namely in situ monitoring instruments and accurateanalyzing methods so as to more timely and accurately graspthe inner activity and state of every potential geologicaldisaster site Moreover this highly advanced easy-to-operatesystem can be considered as a prototype for developinggeological monitoring and warning systems in other regionsthat are prone to geological disasters

AcknowledgmentsThe authors thank all editors for supplying this chanceand thank all viewers for their valuable suggestions and

comments This research was supported by the ldquoWesternlightrdquo Talent Project of the Chinese Academy of Sciences (noY028A11001)

References

[1] S Verma R K Verma A Singh et al Advances in ComputerScience Engineering amp Applications Springer Berlin Germany2012

[2] C Qu H Ye and Z Liu ldquoApplication of WebGIS in seismolog-ical studyrdquo Acta Seismologica Sinica vol 24 no 1 pp 97ndash1062002

[3] KH Kim Y KawataH Kawakata andRGoto ldquoA study on thedevelopment and distribution of WebGIS-based flood hazardmaprdquo Journal of Japan Society for Natural Disaster Science vol23 no 4 pp 539ndash551 2005

[4] F Martinelli and C Meletti ldquoA WebGIS application for render-ing seismic hazard data in Italyrdquo Seismological Research Lettersvol 79 no 1 pp 68ndash78 2008

[5] V Pessina and F Meroni ldquoA WebGis tool for seismic hazardscenarios and risk analysisrdquo Soil Dynamics and EarthquakeEngineering vol 29 no 9 pp 1274ndash1281 2009

[6] F C Yu C Y Chen S C Lin Y C Lin S Y Wu and K WCheung ldquoA web-based decision support system for slopelandhazard warningrdquo Environmental Monitoring and Assessmentvol 127 no 1ndash3 pp 419ndash428 2007

[7] X G Li A M Wang and Z M Wang ldquoStability analysis andmonitoring study of Jijia River landslide based on WebGISrdquoJournal of Coal Science and Engineering vol 16 no 1 pp 41ndash462010

[8] S Gabriele G D Aquila and F ChiaravallotiGeoSpatial VisualAnalytics Springer Dordrecht The Netherlands 2009

[9] LMartino CUlivieriM Jahjah and E Loret ldquoRemote sensingand GIS techniques for natural disaster monitoringrdquo in SpaceTechnologies for the Benefit of Human Society and Earth pp 331ndash382 Springer Dordrecht The Netherlands 2009

[10] H X Lan C D Martin C R Froese et al ldquoA web-based GISfor managing and assessing landslide data for the town of PeaceRiver Canadardquo Natural Hazards and Earth System Science vol9 no 4 pp 1433ndash1443 2009

[11] Z Q Ding and Z H Li Geological Disaster and Prevention inLanzhou City Gansu Science and Technology Press LanzhouChina 2009 (Chinese)

[12] J L Deng ldquoControl problems of grey systemsrdquo Systems andControl Letters vol 1 no 5 pp 288ndash294 1982

[13] J F Chen Z G Shi S H Hong and K S Chen ldquoGreyprediction based particle filter formaneuvering target trackingrdquoProgress in Electromagnetics Research vol 93 pp 237ndash254 2009

[14] E Kayacan B Ulutas and O Kaynak ldquoGrey system theory-based models in time series predictionrdquo Expert Systems withApplications vol 37 no 2 pp 1784ndash1789 2010

[15] Y Y Cao ldquoModeling for grey forecasting of calamities ingeographyrdquo Youthgeogrnphers vol 2 pp 6ndash11 1987

[16] S An J Yan and X Yu ldquoGrey-system studies on agriculturalecoengineering in the Taihu lake area Jiangsu Chinardquo Ecologi-cal Engineering vol 7 no 3 pp 235ndash245 1996

[17] H V Trivedi and J K Singh ldquoApplication of grey system theoryin the development of a runoff prediction modelrdquo BiosystemsEngineering vol 92 no 4 pp 521ndash526 2005

[18] W Gao Computational Methods in Engineering amp Science Spri-nger Berlin Germany 2007


[19] P Lu andM S Rosenbaum ldquoArtificial neural networks and greysystems for the prediction of slope stabilityrdquo Natural Hazardsvol 30 no 3 pp 383ndash398 2003

[20] C C Hsu and C Y Chen ldquoApplications of improved greyprediction model for power demand forecastingrdquo Energy Con-version and Management vol 44 no 14 pp 2241ndash2249 2003

[21] Y F Wang ldquoPredicting stock price using fuzzy grey predictionsystemrdquo Expert Systems with Applications vol 22 no 1 pp 33ndash38 2002

[22] J L Deng ldquoIntroduction to grey system theoryrdquoThe Journal ofGrey System vol 1 no 1 pp 1ndash24 1989

[23] C C Wong B C Lin and C T Cheng ldquoFuzzy trackingmethod with a switching grey prediction for mobile robotrdquo inProceedings of the 10th IEEE International Conference on FuzzySystems pp 103ndash106 Melbourne Australia December 2001

[24] M N Papa V Medina F Ciervo and A Bateman ldquoEstimationof debris flow critical rainfall thresholds by a physically-basedmodelrdquoHydrology and Earth System Sciences Discussions vol 9no 11 pp 12797ndash12824 2012

[25] M Bacchini and A Zannoni ldquoRelations between rainfall andtriggering of debris-flow case study of Cancia (DolomitesNortheastern Italy)rdquoNatural Hazards and Earth System Sciencevol 3 no 1-2 pp 71ndash79 2003

[26] B Y Tan and A Y Duan ldquoStudy on prediction for rainstormdebris flow along mountain district railwayrdquo Journal of NaturalDisasters vol 4 no 2 pp 43ndash52 1995 (Chinese)

[27] H Wu L Shao and D D Lu ldquoThe geological calamity and therainstorm intensity in Lanzhou Cityrdquo Arid Meteorology vol 23no 1 pp 63ndash67 2005 (Chinese)

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ClimatologyJournal of

EcologyInternational Journal of


EarthquakesJournal of


Hindawi Publishing Corporationhttpwwwhindawicom

Applied ampEnvironmentalSoil Science

Volume 2014

Mining


Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal of

Geophysics

OceanographyInternational Journal of


Journal of Computational Environmental SciencesHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal ofPetroleum Engineering


GeochemistryHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Atmospheric SciencesInternational Journal of


OceanographyHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in


MineralogyInternational Journal of


MeteorologyAdvances in

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Paleontology JournalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Geological ResearchJournal of


Geology Advances in


Map of China

Islands of South China Sea

Map of Gansu

Debris flowLandslideYellow river

FaultMain roadCity boundary

ElevationHigh 3670 m

Low 1440 m

Qinghai-Tibetan Plateau

Loess Plateau

N0 10 20 40

(km)

103∘09984000998400998400E 103

∘30

9984000998400998400E 104

∘09984000998400998400E 104

∘30

9984000998400998400E

103∘09984000998400998400E 103

∘30

9984000998400998400E 104

∘09984000998400998400E 104

∘30

9984000998400998400E

37∘09984000998400998400N

36∘309984000998400998400N

36∘09984000998400998400N

37∘09984000998400998400N

36∘309984000998400998400N

36∘09984000998400998400N

Province

Figure 1 The study area landslide sites and debris flow sites with high risk were investigated and supplied by the Lanzhou Bureau of Landand Resources

2 Study Area

Lanzhou city is located in the transitional zone between theQinghai-Tibet Plateau and the Loess Plateau (Figure 1) Inthis area geological conditions are very complex and somedeep and large gullies exist in most parts of the territoryThese geological and geographic backgrounds lead to Lan-zhou being prone to landslides and debris flow disa-sters The latest statistics [11] show that the existing landsl-ide sites (including unstable slopes) number 417 in Lanz-hou In addition in Lan zhou the rainfall from Aprilto September accounts for about 868 of the annual pre-cipitation [11] and always occurs in the form of heavy rainand storms This pattern of rainfall is consistent with thefact that landslide disasters and debris flow always occurduring the months of April to September in Lanzhou Inthe last 5 years (from 2008 to 2012) 16 serious landslidedisasters broke out in this city each landslide disaster causedeconomic losses of more than 01 million CNY 14 among the16 serious landslide disasters were triggered by heavy rainfalland the other 2 landslide disasters were triggered by snowand ice melt detailed presentation is shown in Table 1 At1800 Beijing standard time May 16 2009 the most seriouslandslide disaster occurred at Jiuzhou district in LanzhouThis landslide disaster caused the death of 7 people and the 1person injury of and the direct economic loss of 206 millionCNY was recorded The Jiuzhou landslide belongs to loesslandslide categoryThe length and width of Jiuzhou landslidewere 160 and 40 meters respectively and covered an area ofabout 7500m2 with an average depth of about 40metersThetotal volume was about 62 times 104m3 Figure 2 demonstratesthe front view of the Jiuzhou landslide

3 Landslide and Debris FlowForecasting Method

31 Grey System Forecasting of Landslide Deformation Thegrey system theory was initiated by Deng [12] The conceptof the grey system in its theory and successful applicationis now well known in China It is able to (1) analyze theindeterminate and incomplete data to establish the systematicrelations [13] and (2) forecast time series accurately andthis has been quite a popular subject for researchers bothin the past and at present [14] It assumes that the internalstructure parameters and characteristics of the observedsystem are unknown The system state can be predicted by adifferential equation from the recent historicalmeasurements[13] Although the historical measurements are too complexor chaotic they always contain some governing laws [14]The grey prediction has been widely used in applicationsof geography [15] agriculture [16] runoff prediction [17]displacement prediction of landslide [18] prediction of slopestability [19] power demand [20] stock market [21] and soforthThis theory is also coupled with theWebGIS to forecasttimely deformation in the field of engineering geology suchas the deformation of landslides and host rock in the cavity[7]

The model GM (1 1) one of the grey models is themost widely used in the literature pronounced as ldquogreymodel first order one variablerdquo This model is a time seriesforecasting model The differential equations of the GM (11) model have time varying coefficients In other wordsthe model is renewed as the new data become available tothe prediction model The GM (1 1) model can only beused in positive data sequences [22] In this paper since all




(a) (b)





119911(0)

= 119911(0)

(1) 119911(0)

(2) 119911(0)

(119899)

= 119911(0)

(119896) 119896 = 1 2 119899

(1)

where 119911(0)




119911(1)

= 119911(1)

(1) 119911(1)

(2) 119911(1)

(119899)

= 119911(1)

(119896) 119896 = 1 2 119899

(2)

where

119911(1)

(119896) =

119896

sum

119894=1

119911(0)

(119894) 119896 = 1 2 119899 (3)


119889119911(1)

119889119905

+ 119886119911(1)

= 119887 (4)



(1) = 119911(0)


119886 = [119886

119887] = (119861

119879119861)

minus1

119861119879119884 (5)

where

119861 =

[[[

[

minus119885(1)

(2) 1

minus119885(1)

(3) 1


minus119885(1)

(119899) 1

]]]

]

119884 =

[[[

[

119911(0)

(2)

119911(0)

(3)

sdot sdot sdot

119911(0)

(119899)

]]]

]

(6)

119885(1)

(119896) = 120572119911(1)

(119896) + (1 minus 120572)119911(1)




(1)

(119896 + 1) = [119911(0)

(1) minus

119887

119886

] 119890minus119886119896

+

119887

119886

119896 = 0 1 (7)


(0)

(119896 + 1) = (1)

(119896 + 1) minus (1)

(119896)

= (1 minus 119890minus119886

) [119911(0)

(1) minus

119887

119886

] 119890minus119886119896

(8)





0102030405060708090

100

Disp

lace

men

t (m

m)

Augu

st 10

201

0

Augu

st 20

201

0

Augu

st 30

201

0

Sept

embe

r 10

201

0

Sept

embe

r 20

201

0

Sept

embe

r 30

201

0

Oct

ober

10

201

0

Oct

ober

20

201

0

Oct

ober

30

201

0

Nov

embe

r 10

201

0

Nov

embe

r 20

201

0Fo

reca

sted

Nov

embe

r 30

201

0Fo

reca

sted

Dec

embe

r 10

201

0


0

5

10

15

20

25

30

16 05 1 3 6 24

Accu

mul

ated

rain

fall

(mm

)









4 System Design







(i) Spatial Data










(2) Database


and warning(ISGDMW)

Spatial database

Nonspatialdatabase





Supply data

Upload

Upload


Control module

Solar cell






disaster sites




Monitor


Database

Nonspatialdatabase


Spatial database













Monitoring data

Database ISGDMW


Upload

OperationManager


to outburst


No

Yes



Stag

e one

mon

itor

Stag

e tw

o an

alys

isSt

age t

hree

war

ning






Sensor rain gauge





Grey system theory

Rainfall threshold














References






































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in




(a) (b)





119911(0)

= 119911(0)

(1) 119911(0)

(2) 119911(0)

(119899)

= 119911(0)

(119896) 119896 = 1 2 119899

(1)

where 119911(0)




119911(1)

= 119911(1)

(1) 119911(1)

(2) 119911(1)

(119899)

= 119911(1)

(119896) 119896 = 1 2 119899

(2)

where

119911(1)

(119896) =

119896

sum

119894=1

119911(0)

(119894) 119896 = 1 2 119899 (3)


119889119911(1)

119889119905

+ 119886119911(1)

= 119887 (4)



(1) = 119911(0)


119886 = [119886

119887] = (119861

119879119861)

minus1

119861119879119884 (5)

where

119861 =

[[[

[

minus119885(1)

(2) 1

minus119885(1)

(3) 1


minus119885(1)

(119899) 1

]]]

]

119884 =

[[[

[

119911(0)

(2)

119911(0)

(3)

sdot sdot sdot

119911(0)

(119899)

]]]

]

(6)

119885(1)

(119896) = 120572119911(1)

(119896) + (1 minus 120572)119911(1)




(1)

(119896 + 1) = [119911(0)

(1) minus

119887

119886

] 119890minus119886119896

+

119887

119886

119896 = 0 1 (7)


(0)

(119896 + 1) = (1)

(119896 + 1) minus (1)

(119896)

= (1 minus 119890minus119886

) [119911(0)

(1) minus

119887

119886

] 119890minus119886119896

(8)





0102030405060708090

100

Disp

lace

men

t (m

m)

Augu

st 10

201

0

Augu

st 20

201

0

Augu

st 30

201

0

Sept

embe

r 10

201

0

Sept

embe

r 20

201

0

Sept

embe

r 30

201

0

Oct

ober

10

201

0

Oct

ober

20

201

0

Oct

ober

30

201

0

Nov

embe

r 10

201

0

Nov

embe

r 20

201

0Fo

reca

sted

Nov

embe

r 30

201

0Fo

reca

sted

Dec

embe

r 10

201

0


0

5

10

15

20

25

30

16 05 1 3 6 24

Accu

mul

ated

rain

fall

(mm

)









4 System Design







(i) Spatial Data










(2) Database


and warning(ISGDMW)

Spatial database

Nonspatialdatabase





Supply data

Upload

Upload


Control module

Solar cell






disaster sites




Monitor


Database

Nonspatialdatabase


Spatial database













Monitoring data

Database ISGDMW


Upload

OperationManager


to outburst


No

Yes



Stag

e one

mon

itor

Stag

e tw

o an

alys

isSt

age t

hree

war

ning






Sensor rain gauge





Grey system theory

Rainfall threshold














References






































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in



(1) = 119911(0)


119886 = [119886

119887] = (119861

119879119861)

minus1

119861119879119884 (5)

where

119861 =

[[[

[

minus119885(1)

(2) 1

minus119885(1)

(3) 1


minus119885(1)

(119899) 1

]]]

]

119884 =

[[[

[

119911(0)

(2)

119911(0)

(3)

sdot sdot sdot

119911(0)

(119899)

]]]

]

(6)

119885(1)

(119896) = 120572119911(1)

(119896) + (1 minus 120572)119911(1)




(1)

(119896 + 1) = [119911(0)

(1) minus

119887

119886

] 119890minus119886119896

+

119887

119886

119896 = 0 1 (7)


(0)

(119896 + 1) = (1)

(119896 + 1) minus (1)

(119896)

= (1 minus 119890minus119886

) [119911(0)

(1) minus

119887

119886

] 119890minus119886119896

(8)





0102030405060708090

100

Disp

lace

men

t (m

m)

Augu

st 10

201

0

Augu

st 20

201

0

Augu

st 30

201

0

Sept

embe

r 10

201

0

Sept

embe

r 20

201

0

Sept

embe

r 30

201

0

Oct

ober

10

201

0

Oct

ober

20

201

0

Oct

ober

30

201

0

Nov

embe

r 10

201

0

Nov

embe

r 20

201

0Fo

reca

sted

Nov

embe

r 30

201

0Fo

reca

sted

Dec

embe

r 10

201

0


0

5

10

15

20

25

30

16 05 1 3 6 24

Accu

mul

ated

rain

fall

(mm

)









4 System Design







(i) Spatial Data










(2) Database


and warning(ISGDMW)

Spatial database

Nonspatialdatabase





Supply data

Upload

Upload


Control module

Solar cell






disaster sites




Monitor


Database

Nonspatialdatabase


Spatial database













Monitoring data

Database ISGDMW


Upload

OperationManager


to outburst


No

Yes



Stag

e one

mon

itor

Stag

e tw

o an

alys

isSt

age t

hree

war

ning






Sensor rain gauge





Grey system theory

Rainfall threshold














References






































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in





4 System Design







(i) Spatial Data










(2) Database


and warning(ISGDMW)

Spatial database

Nonspatialdatabase





Supply data

Upload

Upload


Control module

Solar cell






disaster sites




Monitor


Database

Nonspatialdatabase


Spatial database













Monitoring data

Database ISGDMW


Upload

OperationManager


to outburst


No

Yes



Stag

e one

mon

itor

Stag

e tw

o an

alys

isSt

age t

hree

war

ning






Sensor rain gauge





Grey system theory

Rainfall threshold














References






































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in



(2) Database


and warning(ISGDMW)

Spatial database

Nonspatialdatabase





Supply data

Upload

Upload


Control module

Solar cell






disaster sites




Monitor


Database

Nonspatialdatabase


Spatial database













Monitoring data

Database ISGDMW


Upload

OperationManager


to outburst


No

Yes



Stag

e one

mon

itor

Stag

e tw

o an

alys

isSt

age t

hree

war

ning






Sensor rain gauge





Grey system theory

Rainfall threshold














References






































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in



Monitoring data

Database ISGDMW


Upload

OperationManager


to outburst


No

Yes



Stag

e one

mon

itor

Stag

e tw

o an

alys

isSt

age t

hree

war

ning






Sensor rain gauge





Grey system theory

Rainfall threshold














References






































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in







References






































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in




















Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in

research article a webgis-based information system for...

Documents