VEGETATION CHANGES ANALYSIS USING NORMALIZED DIFFERENCE VEGETATION INDEX AND LAND SURFACE TEMPERATUR

Scholarly Research Journal for Interdisciplinary Studies, Online ISSN 2278-8808, SJIF 2021 = 7.380, www.srjis.com PEER REVIEWED & REFEREED JOURNAL, MAR-APR, 2022, VOL- 9/70

VEGETATION CHANGES ANALYSIS USING NORMALIZED DIFFERENCE VEGETATION INDEX AND LAND SURFACE TEMPERATURE IN MEVASI FOREST Vasave Mohan Arjun, Ph. D. Assistant Professor, Department of Geography Arts, Commerce & Science College Taloda District Nandurbar, Maharashtra Email : mohanvasavegeo@gmail.com

Paper Received On: 25 APR 2022 Peer Reviewed On: 30 APR 2022 Published On: 1 MAY 2022

Abstract The present study monitors the interrelationship of Normalized difference vegetation index (NDVI) with land surface temperature (LST) in Mevasi forest in Nandurbar district, Maharashtra, India, using Landsat satellite sensor for the season of 2013 and 2021. Climate change is caused by global warming due to human activities that contribute to greenhouse gas (GHG) emissions. Land-based human activities result in changes in dense vegetation, especially forest stands to land cover with low vegetation density. Remotely sensed multispectral data from Landsat-8 is highly useful in vegetation change analysis based on remote sensing indices and temperature parameters. NDVI (Normalized Difference Vegetation Index)-LST (Land Surface Temperature) relation is essential to understanding the climatological effects on vegetation on regional scales. Thresholdbased classification has been used to realize vegetation change in multi-temporal studies.Similarly, in this study NDVI based classification has been applied to understand the change in the area covered by vegetation and waterbodies. Overall, a weak negative correlation (r = 0.647) was iifound between NDVI-LST and observed that our results based on correlation analysis reaffirmed previous findings for LST-NDVI relations in semiarid regions. ii

ii

iii

ii

ii ii

ii

Scholarly Research Journal's is licensed Based on a work at www.srjis.com

Introduction Forests iiare iisignificant iiin iiclimate iichange iimitigation iimeasures iibecause iithey iiabsorb ii

Carbon iidioxide iiand iitransform iiit iiinto iia iinew iidimension iifor iitrees. iiLoss iiof iiforest iicover,

ii

reduction iiin iiforest iiarea, iior iiforest iidegradation iiresult iiin iigreenhouse iigas iiemissions,

ii

reducing iithe iiforest's iiquality iias iia iirenewable iiresource. (Achmad et al.,2020). Even iion

Dr. Vasave Mohan Arjun

16893

(Pg. 16892-16903)

ii

regional iiscales, iiNDVI ii(Normalized iiDifference iiVegetation iiIndex) iihas iibeen iiused iito iistudy

ii

vegetation iiphenology iichange. iiIdentifying iiclimatological iiand iienvironmental iiinfluences iion

ii

inter-annual iiand iiintra-annual iivariations iiin iivegetation iicover iiis iicritical. iiBecause iivegetation

ii

health iiand iisoil iihumidity iiare iidirectly iirelated, iiNDVI iiand iiLST ii(Land iiSurface

ii

Temperature) iianalyse iivegetation iiconditions iiin iisemiarid iiand iiarid iiregions ii(e.g., iidrought

ii

conditions). iiThe iirelationship iibetween iiNDVI iiand iiLST, iion iieither iihand, iiis iiseasonal, iiwith

ii

a iilarge iiyet iinegative iirelationship iiobserved iiduring iiin iithe iisummer iimonths. (Fenshot et al., 2009). n iicontrast, iia iipositive iicorrelation iiwas iifound iiin iithe iiwinter

ii

months (Sun et al., 2007). Multi-temporal iianalysis iibased iion iiNDVI iiand iiLST iiis iipreferred

ii

to iiunderstand iithe iiadverse iiaffects iiof iidesertification iion iivegetation iicover iibecause iithe

ii

combined iiuse iiof iiboth iiwould iiallow iia iibetter iiunderstanding iiof iichanges iiin iivegetation,

ii

occurring iiin iivarious iiregions. iiFor iieg, iiNDVI iivalues iiwill iibe iisteadier iiand

iialmost iiunvarying iiin iiarid iiregions, iib) iiSemiarid iiregions iiwill iisee iihigher iiNDVI iiwhen iithe ii

temperature iiis iilower, iiand iic) iiIn iitropical iiregions, iiNDVI iivalues iiwill iichange iiassociate iito

ii

temperature ii(Julien iiet iial., ii2011). iiFurthermore, iidirect iialbedo iipresent iivalue iiof iithe iifuture

ii

can iibe iiused iito iihighlight iidifferences iibetween iivarious iicover iitypes. NDVI iihas iialso iibeen iiused iito iiprepare iiland iicover iiclassifications iiat iithe iicontinental

ii

and iiglobal iilevels iisince iimultitemporal iiNDVI iidata iibased iion iiseasonal iiand iiinter ii- iiannual

ii

variations iican iiproduce iivalid iiand iireliable iiresults.(Defries et al.,1994).This iiis iibecause iithese

ii

variations iican iibe iiobserved iidue iito iiclimatic iivariability iior iiactual iichange iiin iiland iicovers ii

ii

In iiaddition, iichanges iiin iivegetation iicover iican iibe iidirectly iiobserved iiby iiNDVI iias iithe

ii

correlation iibetween iivegetation iicover iiand iiNDVI iiis iivery iihigh. iiEven iimoderate iiresolution

ii

satellite iiimagery iieffectively iiunderstands iiand iimonitors iivegetation iicover iidynamics. iiGiven

ii

that iirainfall iidistribution iiand iiother iiclimatic iiparameters iiare iiconstant iiand iiuniformly

ii

distributed iigeographically, iia iimultitemporal iiresearch iibased iion iiNDVI iican iipotentially

ii

identify iivegetation iicover iidegradation iidue iito iitropic iiforces iiover iitime (Jacquin et al., 2010). Land iisurface iitemperature ii(LST) iiis iiconsidered iian iiessential iiparameter iiin iianalyzing

ii

the iiexchange iiof iicomposed iimaterial, iienergy iibalance, iiand iibiophysical iiand iichemical

ii

processes iiof iithe iiland iisurface

ii

and iiLST iivaries iiboth iispatial iiand iitemporally. iiNDVI iiis iiminimum iiin iithe iiadvent iiof iithe

ii

significantly iilower-than-average iiprecipitation iiannually, iiseasonally iidrops iito iilowest iiin iithe

ii

dry iisummer iiand iispatially iithe iiNDVI iihas iivariations, iie.g., iiNDVI iitends iito iibe iihigher iiin iithe

(Tomlinson et al., 2011).In iisemiarid iiregions, iithe iiNDVI

Dr. Vasave Mohan Arjun

16894

(Pg. 16892-16903)

ii

rural iiarea. iiWhereas iiLST-NDVI iirelations iivary iiseasonally, iioverall, iiit iihas iia iiweak iiand

ii

negative iirelationship; iihowever, iiNDVI iirises iiwith iiLST iiin iiwinters. Also, iiin iiall iithree

ii

semiarid iiclimatic iiseasons, iii.e., iisummer, iiwinter, iiand iiautumn, iidaytime iiLSTs iiare iilower iiin

ii

densely iibuilt-up iiareas. (Rasul et al.,2016) iiIn iisemiarid iiclimates, iithe iirelationship iibetween

ii

precipitation iiand iiaverage iiyearly iiNDVI iiis iipositively iicorrelated, iiespecially iiin iithe iigrowing

ii

season. iiSo iiit iiis iiinferred iithat iiprecipitation iican iibe iithe iimain iidriving iifactor iiin iidecreasing

ii

NDVI iias iirecurring iidroughts iiand iiclimate iivariability iihave iicaused iian iiannual iireduction iiin

ii

NDVI, iiespecially iiin iishrublands iiand iicroplands. iiAlthough iiprecipitation iiis iisignificant,

ii

temperature iichanges iican iihave iia iidirect iieffect iion iivegetation, iias iistrong iinegative

ii

correlations iibetween iitemperature iiand iiNDVI iihave iibeen iifound iiin iithese iiregions iiduring iithe

ii

growing iiseason ii(Measho iiet iial., ii2019). iiMoreover, iichanges iiin iiland iiuse ii- iiland iicover iican

ii

have iian iiinfluence iion iiLSTs; iifor iiexample, iiif iiirrigated iiagriculture iiand iiforest iicovers iiare

ii

substituted iiby iibuilt-up iiareas iiover iitime, iiair iiand iiland iiaverage iitemperatures iiwill iiincrease.

In ii ii iicontrast, iiif iibare iisoil iicover iichanges iito iian iiurban iiarea, iithe iiaverage iiLST iiof iithat iiregion ii

can iidecrease; iithus, iivegetation iiand iiurban iisprawl iican iiboth iireduce iiLSTs iion iia iilocal iiscale

ii

even iithough iivegetation iihas iia iicooling iieffect iithrough iitranspiration, iishadows, iiand

ii

rainwater iiaccumulation, iiwhereas iiurban iiregion iican iiplay iithis iirole iibecause iiof iithe iierath

ii

surface ii iiand iicategory iiof iimaterial iithat iimakes iiconvection iimore iieffective iithan iibare iisoil iior

ii

rocky iiareas. ii(Rasul iiet iial.,2017). This iianalysis iiperforms iia iibi- iitemporal iiNDVI iiand iiLST iivegetation iicover iichange

ii

and iiregression iicomparison iiusing iiQGIS ii3.22 iiOpen iiSource iiEnvironment iiin iiMesavi iiForest.

ii

Also, iithis iitimeframe iiwas iichosen iito iiobserve iithe iipost iiscenario iiof iithe iiAsiatic iiLion iicensus

ii

2015, iiwhich iioccurred iiin iithe iipost-monsoon iiperiod iibetween ii2013 iiand ii2021 iiin iithe iimesavi

ii

forest ii. iiWe iibelieve iithat iihigher iiLSTs iihave iia iimajor iiimpact iion iivegetation iicover. iiThis

ii

research iiwill iilook iiat iithe iimost iirecent iichanges iiand iihow iithey iiaffect iivarious iivegetation

ii

cover iitypes, iitaking iiinto iiaccount iidiverse iiconservation iiareas, iiurban iisettlements, iiwater

ii

bodies, iiand iiother iiland iisurface iifeatures.

Study Area The iistudy iiregion iiincludes iitwo iitehsils, iiAkkalkuva iiand iiTaloda, iiin iithe iiNandurbar ii

district iiof iiMaharashtra, iiIndia, iiwith iigeographic iicoordinates iiof ii21° ii30' iiand ii21° ii54' iiN

ii

latitude iiand ii73° ii48' iiand ii74° ii24'E iilongitude. iiIt iicovers iian iiarea iiof ii1358.21 iikm2.

ii

According iito iithe iiKöppen iiGeiger iiclimate iiclassification iimap, iithis iiregion iihas iia iiHot iiSemi-

Dr. Vasave Mohan Arjun

16895

(Pg. 16892-16903)

arid iiclimate ii(Bsh) iiwith iihot, iidry iisummers iiand iimoderate iiwinters ii(Peel iiet iial., ii2007). ii

Additionally, iithis iiregion iiis iipart iiof iithe iiAgro-ecological iiregion iino.6.1, iiwhich iiencompasses

ii

the iiDeccan iiPlateau, iiMaharashtra iiPlains, iiand iiSatpura iimountain iiand iihilly iiregion. iiA iihot-

semi iiarid iiregion iicategorizes iiby iimoderately iideep iiblack iisoils iiand iishallow iidep iisoil (Mandal et al.,2016).

(a) Fig.1 Study area map of Mevasi Forest in Nandurbar District.

Moreover, this study region lies within three major agro-climatic zones of ii

ii

the Western Ghat Zone, all of which are characterised by a dry sub-humid climate.

ii

The iiregion's iivegetation iiis iiclassified iiinto iiTropical iithorn iiforest iiand iipatches iiof iidry

ii

deciduous iiforest. iiThe iiregion's iivegetation iiis iiclassified iiinto iiTropical iithorn iiforests iiand

ii

patches iiof iiDry iideciduous iiforest (Champion et all., 1968).

ii

Precipitation iiis ii712 iimm iifor iiAkkalkuwa iiand ii599.8 iimm iifor iiTaloda iitehsil iiof iiMesavi

ii

forest iiin iiNandurbar iiDistrict. iiAverage iiDaily iiMax. iiAir iitemp. iiAkkalkuva iiis ii32 ii°C iiand iifor

ii

Taloda iiis ii35 ii°C, iiwhereas iithe iiAverage iiDaily iiMin. iiair iitemperature. iiFor iiAkkalkuwa iiis ii21

ii

°C iiand iifor iiTaloda iiis ii26 ii°C. iiAlso, iithe iirelative iihumidity iiof iiAkkalkuwa iiis ii51% iiand iiof

ii

Taloda iiis ii59% ( Indian Metrological Department, 2020).

ii

iiii

ii

The iiAverage iiAnnual

Metholoogy and Database Datasets The iidatasets iiused iiin iithis iistudy iiwere iifrom iithe iiUSGS iiLandsat iisatellite iiLevel-1 iiData ii

Product, iiwhich iiconsists iiof iiraster iiimages iiof iimultispectral iiimage iidata iiin iithe iiform iiDigital

ii

Numbers ii(D.N.) iii.e. iiFor iia iibi-temporal iicomparison, iipixel iivalues iifor iithe iiDecember

ii

months iiof ii2013 iiand ii2021 iiwere iiused. iiBand ii4 ii(Red), ii iiBand ii5 ii(Near-Infrared), iiand iiBand

Dr. Vasave Mohan Arjun

16896

(Pg. 16892-16903)

ii

10 ii(Thermal iiInfrared

ii

Bands ii4 iiand ii5 iihad iia iispatial iiresolution iiof ii30 iim, iiwhereas iiBand ii10 iihad iia iiresolution iiof

ii

100 iim. iiUsing iithe iivector iiboundary iiof iithe iitwo iitehsils iiof iithe iiNandurbar iiDistrict iiindicated

ii

earlier, iirequired iigeo iirectification, iimosaicking, iiand iisubsetting iiwere iiperformed iion iiall

ii

raster iiimages iiused iiin iithis iianalysis.

ii ii

1) iiwere iiutilised iito iicreate iiNDVI iiand iiLST iimaps iiin iithis iistudy.

NDVI Computation Band ii4 iiand iiBand ii5 iiare iiused iifor iicalculating iiNDVI iisince iithe iiratio iiof iithese iibands ii

is iiapplied. iiThe iiNDVI iiis iia iidimensionless iiquantity iiwith iivalues iiranging iifrom ii+1 iito ii-1. iiIt

ii

ranges iifrom ii0 iito ii1, iisignifying iisparse iito iidense iivegetation, iialthough iivalues iiless iithan ii0

ii

indicate iia iicomplete iilack iiof iiflora, iirepresenting iiwater iior iiice. iiIt's iicomputed iiusing iiEq. 𝐍𝐃𝐕𝐈 =

Band ii𝟓−𝐁 iiand ii𝟒

ii

Band ii𝟓+ iiBand ii𝟒

(1)

ii

Band iiRS-GIS iiPlugin iiof iiQGIS ii3.16 iiwas iiused iito iicompute iiinstant iiNDVI iiraster ii

images iias iiit iiconverts iithe iiD.N. iivalues iito iiReflectance iifor iiBand ii4 iiand iiBand ii5 iito iiobtain

ii

NDVI.

LST Computation A iifew iiprocessing iisteps iion iithe iiTIRS iiLevel-1 iidata, iii.e. iiBand ii10, iiare iirequired iito iicompute ii

LST. iiIn iiQGIS ii3.16, iihowever, iiall iiof iithese iiprocedures iiare iicompleted iiautomatically iiusing

ii

RS-GIS iiplugins. a) D.N. iiand iiTOA ii(Top iiof iiAtmosphere) iiradiance ii ii (Lλ) . iiThe iifirst iistep iiis iito iiconvert ii

raw iiD.N. iiinto iiTOA iiradiance, iias iishown iiin iiEq. ii2.

𝐋𝛌 = 𝐌𝐋𝛌 × 𝐐𝐜𝐚𝐥 + 𝐀𝐋𝛌 ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii

(2)

Where iiMLλ iiis iithe iiradiance iimultiplicative iiscaling iifactor iifor iithe iirespective iispectral iiband, ii

ALλ iiis iithe iiradiance iiadditive iiscaling iifactor iifor iithe iirespective iispectral iiband, iiand iiQcal iiis

ii

the iipixel iivalue iii.e., iiD.N. b) Temperature ii(T) iiof iiTOA iiRadiance iito iiAt-Satellite iiBrightness iiEq.3 iishows iiwhat iithe ii

following iistage iiwould iiappear iias.

𝐓𝛌 =

𝐊𝟐 𝐥𝐧

𝐊𝟏 (𝐋𝛌 +𝟏)

− 𝟐𝟕𝟑. 𝟏𝟓 ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii

(3)

Where iiLλ iiis iithe iiradiance, iiK1 iiand iiK2 iiare iiprelaunch iicalibration iiconstants ii(U.S. ii

Geological iiSurvey, ii2016). a) Emissivity iicalculation iibefore iifinal iiLST iicomputation iiAuthor: iiTo iicompute iiLST, iiit ii

is iirequired iito iicalculate iiemissivity ii(e) iias iishown iiin iiEq. ii4.

Dr. Vasave Mohan Arjun

16897

(Pg. 16892-16903)

𝐞 = 𝟎. 𝟎𝟎𝟒𝐏𝐯 + 𝟎. 𝟗𝟖𝟔 ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii

(4)

Where iiPv iiis iithe iivegetation iiproportion iiand iiis iicalculated iiwith iithe iihelp iiof iiscaled iiNDVI ii

(by iiusing iithe iiNDVI iiobtained iiearlier) iias iishown iiin iiEq. ii5. 𝐍𝐃𝐕𝐈−𝐍𝐃𝐕𝐈𝐦𝐢𝐧

𝐏𝐯 = [𝐍𝐃𝐕𝐈

𝐦𝐚𝐱 −𝐍𝐃𝐕𝐈𝐦𝐢𝐧

]

𝟐

(5)

ii ii ii

Where iithe iiNDVI iiis iicomputed iiearlier iiper iipixel. iiWhile, iiNDVImin iiand iiNDVImax are iithe ii

ii

minimum iiand iimaximum iiNDVI, iirespectively. iiThe iiequation iiportion iiin iithe iisquared

ii

brackets iiis iialso iicalled ii'scaled iiNDVI' ii(Carlson ii& iiRipley, ii1997). c) LST Calculation 𝐋𝐒𝐓 =

𝟏+(

𝐓𝛌 𝛌×𝐓𝛌 ) 𝛒

(6)

Lne

Where, Tλ is the at-satellite brightness temperature, λ is the wavelength of the emitted radiance, ρ = h×c/j (h is Planck’s constant i.e. 6.62607015×10-34 Js , c is velocity of light i.e. 2.99 x 108 m/s and j is the Boltzmann constant i.e. 1.380649×10-23 J K-1) and as mentioned earlier the emissivity (e) computed using Eq. 4 will be further used to calculate the final LST (Artis & Carnahan, 1982). NDVI Derived Vegetation Cover Thresholding iiNDVI iivalues iiprepare iithe iivegetation iicover iimaps. iiThe iithreshold ii

values iiof iithe iicover iitypes iiare iiapproximately iibased iion iithe iireference iistudies iias iimentioned

ii

in iiTable ii1. iiThe iibare iiLand iithreshold iiwas iibased iion iitwo iistudies, iione iidirectly iiand iithe

ii

other iiindirectly iibased iion iithe iiminimum iithreshold iivalue iiof iithe iicrop. iiSimilarly, iithe iisparse

ii

vegetation iithreshold iiwas iiindirectly iibased iion iithe iiminimum iithreshold iivalues iiof iicrops.

Table ii1. iiNDVI iithreshold iivalues iiof iidifferent iicover iitypes. Sr iiNo

NDVI ii iiValue iiThreshold ii

Class iiType

Reference

0

Water iibodies

-0.046

(Bisrat ii& iiBerhanu, ii2018), (Dalezios iiet iial., ii2001)

1

Bare iiLand

0.25

(Ding iiet iial., ii(Thorat et iial., ii2015)

2

Low iiVegetation

0.35

(Thorat iiet iial., ii2015)

3

Moderate iiVegetation

0.5

(Bisrat ii& iiBerhanu, ii2018), (Dalezios iiet iial., ii2001)

4

Dense ii iiVegetation

1

(Dalezios iiet iial., ii2001)

ii2016)

Dr. Vasave Mohan Arjun

16898

(Pg. 16892-16903)

Result and Discussion As iishown iiin iiFig.1 iiand fig.2, iiit iiis iiconfirmed iiclearly iithat iithe ii2013 iiNDVI iivalues ii

are iisignificantly iilower iithan ii2021 iiNDVI iivalues, iias iithe iimean iiNDVI iivalue iifor ii2013 iiwas

ii

found iito iibe ii0.39 ii(±0.12 iiS.D.) iiin iicontrast iito iithe ii2021 iimean iiNDVI iivalue iiwhich iiwas ii0.43

ii

(±0.12 iiS.D.). iiFor ii2013, iithe iiminimum iiand iimaximum iiNDVI iivalues iiwere ii- ii0.75 iiand ii0.81,

ii

respectively. iiFor ii2021, iithe iiminimum iiand iimaximum iiNDVI iivalues iiwere ii– ii0.53 iiand ii0.80,

ii

respectively. iiIn iithe iicentral iipart iiof iithe iistudy iiarea iiwhere iithe iiKhardi iiand iiVarkhedi ii iiRiver

ii

Basin iilies, iithere iiwas iia iiclear iidifference iiin iiNDVI, iisuggesting iithe iisignificantly iivarying

ii

vegetation iicover iiwhile iicomparing iithe iitwo iiyears.

Dr. Vasave Mohan Arjun

16899

(Pg. 16892-16903)

Fig.3 (a). iiLST iiMap iiof ii2013 & 2021 iiiifor Mevasi iiMesavi iiForest iiin iiNandurbar

Mevasi Forest: Corelation between NDV and LST (2013)

40.00 35.00

30.00

y = -2.3039x + 25.096 R² = 0.0337

LST (ºC)

25.00 20.00 LST

15.00

Linear (LST)

10.00 5.00 0.00 0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

NDVI

0.80

Dr. Vasave Mohan Arjun

16900

(Pg. 16892-16903)

40.00

Mevasi Forest: Corelation between NDV and LST (2021)

35.00

y = -6.5222x + 30.666 R² = -0.1463

30.00

LST (ºC)

25.00 20.00

LST

15.00

Linear (LST) 10.00 5.00

0.00 0.00

0.20

0.40

0.60

0.80

NDVI ii

Fig. ii4 iiLST- iiNDVI iiScatterplot iifor iicompering ii2013 iiand ii2021 iiValues.

Also, iithe iiLST iimaps, iias iishown iiin fig.3 iiand fig.4 iireveal iia iisignificant iidifference iiin ii

LSTs iiwhile iicomparing iithe iitwo iiyears. iiIn ii2013, iithe iimean iiLST iiof iithe iientire iiarea iiwas

ii

24.39 ii°C ii(±1.9 iiS.D.), iiwhereas iiin ii2021, iithe iimean iiLST iiwas ii28.3 ii°C ii(±2.5 iiS.D.). iiThis

ii

clearly iisuggests iithat ii2013 iiyear iimust iihave iihigher iitemperatures iieven iiin iiDecember iias iiit

ii

is iialmost iithe iionset iiof iiwinter iiin iithis iiregion. In Comparison, 2013 iiwitnessed iia iistark iidifference iiin iiLSTs iibetween iidifferent iiland

ii

cover iitypes iie.g. iiwater iibodies iiand iibarren iiareas iiin iithe iinorth, iiin iicontrast iito ii20121, iiwhere

ii

the southern central

ii

differences iiwere iinot iisignificant iias ii2013 iiLSTs. Our iiresults iiagree iiwith iia iistudy iithat

ii

implies iistrong iipositive iicorrelations iibetween iiLST iiand iiNDVI iiare iionly iiwitnessed iiin iithe

ii

warm iimonths, iiincluding iisummer. iiWhen iiapproaching iithe iionset iiof iiwinter ii(i.e., iiDecember

ii

in iiour iistudy iiarea), iithe iichange iifrom iinegative iito iipositive iicorrelation iistarts (Sun et all.,

ii

part iihad iihigher iitemperatures iithan iithe north,

ii

and iitemperature

2007). ii As iishown iiin fig.5. iiAn iioverall iiweak iipositive iicorrelation ii(r ii= ii0.647) iiwas iifound ii

for iiboth iiyears. iiIn iiaddition, iithere iiwas iian iieven iiweaker iipositive iicorrelation ii(0.146) iifound

ii

in ii2021 iibetween iiLST- NDVI iivalues iicompared iito ii2013, iiwhere iialthough iiit iiwas iia iiweak

ii

negative ii(0.033), iiit iiwas iistill iistronger iithan iiin ii2021. iiMoreover, iias iishown iiin fig. 5. iiscatter

ii

plot, iithe iiNDVI iivalues iiin ii2013 iiare iialmost iinormally iidistributed iicompared iito ii2021, iiwhich

ii

is iislightly iipositively iiskewed iiand iishows iiit iihas iirelatively iilower iipositive iiNDVI iivalues

16901

Dr. Vasave Mohan Arjun (Pg. 16892-16903)

ii

than ii2013. iiThis iialso iiconfirms iithe iiresult iiof ii2021 iiLSTs iiwhich iishows iicooler iitemperatures

ii

than iiin ii2013. When iicomparing iiFigures ii6 iiand ii7, iithe iivariation iiin iivegetation iicover iitypes,

ii

particularly iiin iithe iidense iivegetation iiclass, iiis iisignificant. iiSparse iiand iiModerate iivegetation-

covered more area in 2013 than in 2021. iiDense vegetation patches are few in the ii

ii

2013 iivegetation iicover iimap. iiMost iiimportantly, iidense iivegetation iicover iipatches iiwere

ii

almost iiabsent iiin ii2015 iiin iithe iiKhardi iiRiver iiBasin iiarea. ii Table No .3 Mevasi iiForest: iiArea iiin iiVegetation iichanges Sr. iiNo

Class

Area iiin iisq iikm (2013)

Area iiin iisq iikm (2021)

% (2013)

% (2021)

1

Water iibodies

10.79

9.79

0.79

0.72

2

Bare iiLand

22.66

40.59

1.67

2.99

3

Low iiVegetation

435.73

634.13

32.08

46.69

4

Moderate iiVegetation

793.86

607.85

58.45

44.75

5

Dense ii iiVegetation

95.17

65.84

7.01

4.85

1358.21

100

Total

Table, 3 presents comparison statistics for the area.As per Table 2, the ii

ii

Dense vegetation cover in 2021 was almost 2.16 % higher than in 2013, which

ii

can be because the LSTs in 2013 were higher than in 2021, and higher LSTs can

ii

be due to dryness or drought-like conditions in that year. Therefore 2013 has less

ii

dense vegetation cover than 2021. Water cover was lower by almost 20.58 sq. km

ii

(0.07%) in 2013 than in 2021. Additionally, Bare soil area had been reduced from

ii

2013 to 2021 by more than 1.32 %. This shows clearly significant variations in

ii

vegetation cover between the two years.

ii

Conclusion i

ii

Our results reveal that strong positive relationships between LST and NDVI ii

ii

can only be seen in decamped months, i.e., in the winter. Positive relationships are ii

ii

found in our climatic zone, as well as other zones throughout the world. Another

ii

noteworthy finding is that the percentage of water bodies was much lower in 2013

ii

than in 2021, implying that decreasing precipitation has an impact on NDVI

ii

values, as well as irrigation in semiarid areas like our study area.

ii

To overcome the limitations of the LST-NDVI relationship for vegetation change ii

ii

study, a precipitation and irrigation scenario can be included to better understand ii

ii

16902

Dr. Vasave Mohan Arjun (Pg. 16892-16903)

ii

the NDVI relationship. For similar threshold-based vegetation cover classification,

ii

we recommend using several indices like the Enhanced Vegetation Index (EVI) or

ii

Perpendicular Vegetation Index (PVI), Vegetation Health Index (VHI), and

ii

Normalized Differences Water Index (NDWI).

ii

Refrences Artis, iiD. iiA., ii& iiCarnahan, iiW. iiH. ii(1982). iiSurvey iiof iiemissivity iivariability iiin iithermography iiof iiurban iiareas. iiRemote iiSensing iiof iiEnvironment, ii12(4), ii313– 329. iihttps://doi.org/10.1016/0034- ii4257(82)90043-8. Bisrat, iiE., ii& iiBerhanu, iiB. ii(2018). iiJOURNAL iiOF iiNATURAL iiRESOURCES iiAND iiDEVELOPMENT iiIdentification iiof iiSurface iiWater iiStoring iiSites iiUsing iiTopographic iiWetness iiIndex ii(TWI) iiand iiNormalized iiDifference iiVegetation iiIndex ii(NDVI) iiArticle iihistory. iiJournal iiof iiNatural iiResources iiand iiDevelopment, ii08, ii91–100. iihttps://doi.org/10.5027/jnrd.v8i0.09. ii ii ii Carlson, iiT. iiN., ii& iiRipley, iiD. iiA. ii(1997). iiThe iirelation iibetween iiNDVI, iifractional iivegetation iicover, iiand iileaf iiarea iiindex. iiRemote iiSensing iiof iiEnvironment, ii62(3), ii241–252. iihttps://doi.org/10.1016/S0034-4257(97)00104-1 iiChampion, iiH. iiG., ii& iiSeth, iiS. iiK. ii(1968). iiA iiRevised iiSurvey iiof iithe iiForest iiTypes iiof iiIndia. iiGovernment iiof iiIndia. Dalezios, iiN. iiR., iiDomenikiotis, iiC., iiLoukas, iiA., iiTzortzios, iiS. iiT., ii& iiKalaitzidis, iiC. ii(2001). iiCotton iiyield iiestimation iibased iion iiNOAA/AVHRR iiproduced iiNDVI. iiPhysics iiand iiChemistry iiof iithe iiEarth, iiPart iiB: iiHydrology, iiOceans, iiand iiAtmosphere, ii26(3), ii247–251. iihttps://doi.org/10.1016/S1464-1909 ii(00) ii00247-1. Defries, R. S., & Townshend, J. R. (1994). Ndvi-Derived Land Cover Classifications At a Global Scale. International Journal of Remote Sensing, 15(17), 3567–3586. https://doi.org/10.1080/01431169408954345.. Ding, iiY., iiZheng, iiX., iiZhao, iiK., iiXin, iiX., ii& iiLiu, iiH. ii(2016). iiQuantifying iithe iiimpact iiof iiNDVIsoil iidetermination iimethods iiand iiNDVIsoil iivariability iion iithe iiestimation iiof iifractional iivegetation iicover iiin iiNortheast iiChina. iiRemote iiSensing,8 ii(1), ii1– ii15. iihttps://doi.org/10.3390/rs8010029. Fensholt, R., Rasmussen, K., Nielsen, T. T., & Mbow, C. (2009). Evaluation of earth observation based long term vegetation trends - Intercomparing NDVI time series trend analysis consistency of Sahel from AVHRR GIMMS, Terra MODIS and SPOT VGT data. Remote Sensing of Environment, 113 (9),1886-1998. Indian iiMeteorological iiDepartment. ii(2010). iiClimatological iiTables iiof iiObservations iiin iiIndia, ii1981-2010. ii893. iihttp://www.imdpune.gov.in/library/public/1981-2010 iiCLIM iiNORMALS ii(STATWISE).pdf. Jacquin, A., Sheeren, D., & Lacombe, J. P. (2010). Vegetation cover degradation assessment in Madagascar savanna based on trend analysis of MODIS NDVI time series. International Journal of Applied Earth Observation and Geoinformation,12(SUPPL.. Peel, iiM. iiC., iiFinlayson, iiB. iiL., ii& iiMcMahon, iiT. iiA. ii(2007). iiUpdated iiworld iimap iiof iithe iiKöppen-Geiger iiclimate iiclassification. iiHydrology iiand iiEarth iiSystem iiSciences, ii11(5), ii1633–1644. iihttps://doi.org/10.5194/hess-11-1633-2007. Copyright © 2022, Scholarly Research Journal for Interdisciplinary Studies

Dr. Vasave Mohan Arjun

16903

(Pg. 16892-16903) Mandal, D. K., Mandal, C., & Singh, S. K. (2016). Delineating Agro-Ecological Regions. National Bureau of Soil Survey and Land Use Planning, 1–8. Rasul, A., Balzter, H., & Smith, C. (2016). Diurnal and seasonal variation of surface Urban Cool and Heat Islands in the semiarid city of Erbil, Iraq. Climate, 4(3). https://doi.org/10.3390/cli4030042. Rasul, A., Balzter, H., & Smith, C. (2017). Applying a normalized ratio scale technique to assess influences of urban expansion on land surface temperature of the semi-arid city of Erbil. International Journal of Remote Sensing, 38(13), 3960–3980. https://doi.org/10.1080/01431161.2017.1312030. Sun, D., & Kafatos, M. (2007). Note on the NDVI-LST relationship and the use of temperature-related drought indices over North America. Geophysical Research Letters, 34(24), 1–4. https://doi.org/10.1029/2007GL031485. Thorat, iiS. iiS., iiRajendra, iiY. iiD., iiKale, iiK. iiV., ii& iiMehrotra, iiS. iiC. ii(2015). iiEstimation iiof iiCrop iiand iiForest iiAreas iiusing iiExpert iiSystem iibased iiKnowledge iiClassifier iiApproach iifor iiAurangabad iiDistrict. iiInternational iiJournal iiof iiComputer iiApplications, ii121(23), ii43–46. iihttps://doi.org/10.5120/21845-515.3. U.S. iiGeological iiSurvey. ii(2016). iiLandsat ii8 iiData iiUsers iiHandbook. iiNasa, ii8(June), ii97. iihttps://landsat.usgs.gov/documents/Landsat8DataUsersHand iibook.pdf.

Turn static files into dynamic content formats.

Create a flipbook