development of spectral library of mangrove native species of...

Indian Journal of Geo Marine Sciences

Vol. 49 (5), May 2020, pp. 748-756

Development of spectral library of mangrove native species of the

Muthupet lagoon, Tamil Nadu, India using field spectroscopic instrument

A Selvaraja,$, S Saravanan*,a,† & N Ranib aDepartment of Civil Engineering, National Institute of Technology, Tiruchirappalli, Tamil Nadu – 620015, India

bGeological Survey of India, Kolkata, West Bengal – 700 016, India

[E-mail: †[email protected]; [email protected]]

Received 12 November 2018; revised 12 January 2019

In the recent trends, the spectro-radiometer technique has proved its significance in the development of spectral library

of earth’s surface materials to support the hyperspectral image interpretation and analysis. To enhance the efficiency of

hyperspectral image classification, acquisition of spectral signature must be carried out. A practical study was carried out on

Muthupet mangrove forest, Tamil Nadu, India. According to remote sensing data, the selected study shows as contiguous

lagoon, and Avicennia marina (95 % occupancy) is the majority community of the study area followed by Acanthus

ilicifolius, and Aegiceras corniculatum. It is the first attempt to develop the spectral library for mangrove true species, and

mangrove associates for the Muthupet lagoon, Tamil Nadu, India. With detailed field visits, by using the analytic spectral

device (ASD), the spectral library built for prominent 6 true mangrove species such as Acanthus ilicifolius, Aegiceras

corniculatum, Avicennia marina, Excoecaria agallocha, Rhizophora apiculata and Rhizophora mucronata) in connection to

that 5 mangrove associates Salicornia brachiata, Sesuvium portulacastrum, Suaeda martima, Suaeda monoica, and Azims

tetracantha. The reflectance and absorption of each species was observed, and built a spectral library of mangroves species.

It is found that Acanthus ilicifolius provides high reflectance than others. This spectral library can be useful for mangroves

species identification using hyperspectral data.

[Keywords: Avicennia marina, Field spectro-radiometer, First derivative of reflectance, Mangrove associates,

Mangrove species]

Introduction

The Mangrove forests are salt-tolerant vegetation

community of woody hydrophytes. It originates from

tropical and subtropical zone of ecosystem, which

covers various features such as estuaries, lagoons,

creeks and mudflats, creeks1-3. The forests are dense

with spreading trees since each species has different

roots structure such as knee roots, pneumatophores

and stilt roots4. The remote sensing is the most

important foundation of spatial consequence of the

earth’s surface5. The hyperspectral remote sensing

system is unique and provides the fine spectral

resolution of objects in many narrow bands6 it can be

able to measure the quantitative information over

vegetation with physicochemical status7. The spectral

measurement using hyperspectral sensor is helpful in

classifying the materials in ecosystem characteristics8.

The effects of environmental processes and trouble on

the spatio-temporal distribution of vegetation

communities are significant aspects, and concern

organizations are responsible to protect the natural

zone level9. Many authors made use of reference

spectral libraries for mineral mapping10-14. This idea

can be inspired recently for the mapping of vegetation

using in situ reflectance measurements)15-20.

The species identification using multispectral

bands is not good enough due to absence of much

absorption information for various crops. The better-

quality spectral discrimination scheme is not

available21, therefore suitable technique is required in

classifying similar class on the data either using

training data extracted from the image or spectral

library22. Many spectral libraries (soil, water,

vegetation are available for user23, those are

supporting for visual interpretation and upgraded-in to

current commercial software. However, existing

spectral library did not have some reference spectra

for mangrove species. To develop the reference

spectral library for mangrove species, an extensive

field work was carried out and incorporated with

unknown mangrove species22. It has been observed

that taking measurements on land is uncomfortable

and difficult to use electronic equipment in lagoon.

The measurement of leaf reflectance in 531 nm can

vary within minutes after shifting dark to full light in

field conditions without detachment27. Collection of

SELVARAJ et al.: DEVELOPMENT OF SPECTRAL LIBRARY OF MANGROVE NATIVE SPECIES USING ASD

749

mangrove leaves from the forest is a common practice

for spectral measurements as filed access is very

complicated28-30.

The spectro-radiometer is a special kind of optical

instrument, which is designed for general purpose,

and very useful in many application, particularly for

remote sensing application. It acquires the spectra

from blue-green-red, infrared and short wave infrared

portion and used for various application. Various

studies conducted for hyperspectral measurement for

vegetation species identification31-34. The analytical

spectral device consists of more than 2000 channel

with a range of 3.5–2.5 µm26,31. The non imaging

field instrument capable of providing more than

100 individual bands, and provided contiguous

information about the earth surface35. It acquires good

spectral radiance from field, and retrieves the absolute

reflectance of earth’s surface targets. In ancient years,

the field spectro-radiometers exploited to study about

colour vision36. Currently two types of spectro-

radiometer available in the market are (1) small, light

devices that are designed to operate in the visible near

infrared with considerable signal-to-noise ratio and

(2) larger, heavier devices which are responsive to the

visible, infrared, near infrared, and shortwave infrared

of spectrum. A ground based investigation is being

carried out by acquiring the spectral profile from the

ASD in fresh leaf scale37. To enhance the

hyperspectral data classification, a reference spectral

library is needed, but the available spectral library like

United states geological survey (USGS) does not have

the own unique library for native mangroves species.

The main task of article is to develop a spectral

library for mangrove species and mangrove associates

from non imaging instrument.

Materials and Methods

Study area

The Tamil Nadu shoreline is located on the

southeast coast of Indian peninsula, and it is about

1,076 km long. The cauvery delta is located in the

region of Tamil Nadu state in southern India. It covers

the lower reaches of Cauvery and its delta. The delta

regions covers the Muthupet lagoon area (Latitude:

10°18’13” to 10°20’71” N and Longitude: 79°30’90”

to 79°34’87” E; Figure 1) along the coastal zone of

the Bay of Bengal and Palk strait, is selected for

mangrove spectral library generation. The stream

drained by the tributaries of the Cauvery river

such as the Paminiyar, Koraiyar, Kandankurichanar,

Kilathangiyar and Marakkakoraiyar. This forest

receives freshwater mostly through the northeast

monsoon season from October to November24.

The area of Muthupet mangrove wetland is about

12,000 ha, and it is divided into six reserve forests. It

consists of two large lagoons, which are contiguous

and about 1,700 ha in area. The botanical survey

reviewed that there are total 8 numbers of true

mangrove species present in the Muthupet mangroves.

Among them three species namely: Ceriops decandra,

Rhizophora apicualta and Rhizophora mucronata

were reintroduced recently24. The current status of this

study area is that there were degraded of mangroves,

and only six mangrove species available. The

Mangrove associates such as Suaeda maritima,

Suaeda monica and Salicornia brachiata are also

found in the in landward margin wetlands24.

The mangrove leaves were picked and packed with

plastic covers and arranged for measurement of

optical properties of leaves25. The spectral acquisition

can be done across narrow bands in the region of

3.5–10 µm, 10–1.83 µm and 1.83–2.50 µm26. The

laboratory spectra data carried out for the 6 true

mangroves species and 5 Mangrove associate during

July 29, 2018. Before collecting all the species,

authors frequently visited the Muthupet mangrove

forest, and observation is made about the leaves

structure and root of each species, location and

abundance of mangrove species and associates. The

following observations are made during field visit.

The distribution of species is shown in Figure 2. 1)

Seaward zone occupied by Avicennia marina as

dominant, followed by Aegiceras corniculatum; 2)

The mid zone is mixing of Avicennia marina with

small patches of Rhizophora apiculata, and

Rhizophora mcronta and 3) The terrestrial zone is

occupied by Souada zone with small patches of

Prosopis juliflora.

The field photographs which exhibit the canopy

status of mangrove as shown in Figure 3. The Spectral

library is developed for all the healthy mangrove

species. The view Spec used to build the spectral

profile of those species.

The view spec tool is also permits the larger

number of files and visualize graphically. It has

provision such as Log 1/R (1/T), 1st Derivative, 2nd

Derivative and Lambda Integration. The GARMIN

GPS Etrex-10 used to collect the location of each

species as shown in Table 1. The specification of GPS

is 1000 way points, and 50 routes. The First derivative

INDIAN J GEO-MAR SCI, VOL 49, NO 05, MAY 2020

750

Fig. 1 — Location of study area

Table 1 — List of mangrove true species and its associates collected from Muthupet mangrove forest

S. No Coordinates Name of true mangrove species

1 N 10˚ 21' 15.53'' E 79˚ 31' 51.28'' Avicennia marina

2 N 10˚ 21' 15.23'' E 79˚ 31' 51.38'' Aegiceras corniculatam

3 N 10˚ 21' 09.89'' E 79˚ 32' 11.61'' Rhizophora mucronata

4 N 10˚21' 09.81'' E 79˚ 31'39.61'' Excoeceria agallocha

5 N 10˚21' 14.08'' E 79˚ 32'05.44'' Acanthus ilicifolius

6 N 10˚21' 11.28'' E 79˚21' 28.38'' Rhizhopora apiculta

Coordinates Name of mangrove associate

7 N 10˚ 21' 20.56'' E 79˚ 31' 51.09'' Suadea Martima

8 N 10˚ 21' 16.54'' E 79˚ 31' 49.22'' Salicoronia brachiata

9 N 10˚ 19' 22.57'' E 79˚ 32' 24.42'' Sesuvium portulaccustrum

10 N 10˚ 19' 23.06'' E 79˚ 32' 24.37'' Azhima tetracantha

11 N 10˚ 19' 23.04'' E 79˚ 32' 24.32'' Suadea monoica


751

of the reflectance of mangrove species can be built

through view Spec tool. Finally, the spectral signatures

were exported in the Environmental visualization (ENVI

4.7) software for building of the library. In this study, all

the spectral signatures imported into the spectral library

tool, and extract the signatures curves for the mangroves

and its associates. The flow chart for over all

methodology is show in Figure 4. The Physcio-chemical

parameter of true mangrove species, and mangrove

associates are shown in Figure 5 and 6 respectively.

Rhizophora apiculata and Rhizophora mucronata have

introduced recently by the Tamilnadu forest department,

the leaves of those Rhizophoracy family are very thicker

and dark green in color.

Fig. 2 — Typical zone of mangrove species distribution in Muthupet mangrove forest

(Source: https://www.pngwing.com/en/free-png-bogao)

Fig. 3 — Field photographs showing canopy status of mangroves


752

Results and Discussion

The spectral library was generated for True

mangrove species and its associates. The absorption

of each species as mentioned in Table 2. Among the

mangrove species, the Excoeceria agallocha has the

high reflectance peak, i.e. 82 % at 1086 nm across

very near infrared of electromagnetic spectrum, the

reflectance is maximum between 672 nm and 1273

nm wavelength of infrared spectrum. Red edge for the

Avecenniya marina is originated at the 672 nm. The

reflectance is maximum 33 % at 548 nm. The

maximum absorption occurs at 1.43 µm and the

reflectance is very low at Short wave infrared region

i.e. 17 % of reflectance at 1672 nm.

Acanthus ilicifolius has maximum reflectance at

1070 nm in the near infrared regions; however,

reflectance is peak at 552 nm of green wave length.

Absorption peak is taking place at the wavelength of

1432 nm. The Aegiceras corniculatum has high

reflectance in 550, moreover the reflectance is high

between the wavelengths of 701 nm to 1263 nm of near

infrared of EMR, the absorption peak at the wavelength

of 1430 nm of near infra-red. Both Rhizophora

apiculata and Rhizophora mucronata almost behave

identical at visible, near Infra red, and short wavelength

infrared region. The Rhizhopora apicualta met

maximum peak at the wavelength of 1078 nm of

infrared of electromagnetic spectrum. Similarly

Fig. 4 — Flow chart for methodology

Table 2 — Absorption details of mangrove species

S. No. Species name Absorption (nm)

1 Avicennia marina 400, 500, 950, 1100, 1400, 1900

2 Aegiceras corniculatum 401, 505, 955, 1101, 1401, 1900

3 Rhizophora apliculata 400, 500, 949, 1102, 1400, 1900

4 Rhizophora mcronota 400, 499, 950, 1100, 1400, 1900

5 Excoeceria agallocha 400, 498, 951, 1100, 1399, 1900

6 Acanthus ilicifolius 402, 500, 950, 1100, 1400, 1900


753

Rhizophora mucronata has maximum reflectance at

548 nm of visible spectrum, and attained maximum

reflectance peak of 75 % at the wavelength of 1075 nm

of near infrared regions, since presence of water

absorption bands, the reflectance is very low, and the

absorption peak at 1432 nm of short wavelength

infrared of the EMR. The Avicennia marina is

dominant occupancy in the study area, which exhibit

maximum reflectance about 72 % at 1088 nm in NIR,

The reflectance is high between the wavelength of

679 nm to 1268 nm, and reflectance 27 % in 555 nm of

visible spectrum of EMR, the absorption is peak at

1434 nm of SWIR. Among the mangrove associates,

Sensuvium portulacastrm provides maximum peak

Fig. 5 — Physio-chemical parameter of mangrove species

Fig. 6 — Physio-chemical parameter of mangrove associates


754

at NIR, followed by Suada zones, Salicornia brachiata,

and Salvadoraceae. Despite mangrove associates are

not salt tolerant, most of patches are found to be

degraded condition and it is continue to support

to mangroves true species, the plot of reflectance

of all species as shown in Figure 7.

First derivative of reflectance of those mangrove

species is shown in Figure 8, which is provided

the peak reflectance and absorption details across

wavelength. To understand the reflectance of each

species, the offset has been set (Figure 9). When

compared to existing spectral library, the reflectance of

mangrove species can vary with respect to

environmental condition, generally mangroves are

distributed in the coastal zone, since there is lack of

fresh water, hyper salinity and degraded conditions,

and the reflectance is almost low when compared to

reflectance of crops in various portion of EMR.

The Prosopis juliflora provides high reflectance than

the mangroves.

Fig .7 — Plot of Spectral library of mangrove true species and its associates

Fig. 8 — First derivative of reflectance of mangrove species and its associates

Fig. 9 — Plot of reflectance of mangrove species and its

associates


755

Conclusion

In this article, The spectral library generation for

mangroves is carried out, which is highly required for a

spectral matching techniques, existing spectral library

like USGS does not provide the spectra library for

mangrove species since canopy structure is very much

depend on natural parameter like salinity, and soil

condition of zone, in advancement of hyperspectral

technology, it is first attempted to develop the spectral

library for true mangrove species and its associate in

Muthupet mangrove forest. 25 numbers of samples are

utilized to collect the reflectance of true species. The

spectral library built for prominent 8 true mangrove

species. (Acanthus ilicifolius, Aegiceras corniculatum,

Avicennia marina, Excoecaria aagallocha, Rhizophora

apiculata and Rhizophora mucronata) in connection to

that 5 mangrove associates (Salicornia brachiata,

Sesuvium portulacastrum, Suaeda martima, Suaeda

monoica, Azims tetracantha). The near infra red

portion of EMR plays vital role for the mangroves

species rather than the short wave infrared region.

Among the mangroves species, Excoeceria agallocha

has high reflectance across very near-infra red of

electromagnetic spectrum i.e. 82 % of reflectance at

1086 nm and the reflectance is maximum between 672

nm and 1273 nm wavelength of infrared spectrum. The

Avicennia marina is dominant occupancy in the study

area, which exhibit the maximum reflectance about 72

% at 1088 nm in NIR. The Aegiceras corniculatam

provides high reflectance between the wavelengths of

701 nm to 1263 nm of Near Infrared of EMR.

Acanthus ilicifolius has maximum reflectance at

1070 nm in the NIR, the Rhizophora apiculata and

Rhizophora mucronata almost behave identical at

visible, near infra red, and short wavelength infrared

region. In connection to that among the mangrove

associates, Sensuvium portulacastrm provided

maximum peak at NIR, followed by Suada zones,

Salicornia brachiata, and Salvadoraceae. Thus, the

spectral library was built for all the mangrove species.

The reflectance and absorption of each species

was observed. Among the mangrove species, the

Acanthus ilifolius provides high reflectance. Finally

this spectral library can be useful for spectral matching

techniques in Hyperspectral approach, and image

classification can be improved with higher accuracy.

Acknowledgements

We would like to express our sincere thanks to

Geological Survey of India, Bengaluru, Hyperspectral

division for sparing the ASD Field Spectro-

radiometer for the collection of spectral signatures.

We extended sincere thanks to Forest Department,

Tamil Nadu, India for their supporting field visit and

species collection.

Conflicts of Interest

The authors declare no conflict of interest.

Author Contributions

AS collected the data, designed the method of

study and wrote the major part of articles. SS

visualized the method and wrote part of the

manuscript and also performed the revision and NR

prepared the method for spectral data collection and

visualized the revision.

References 1 Kathiresan K & Bingham B L, Biology of Mangroves and

Mangrove Ecosystems, Adv Mar Bio, 40 (2001) 81–251.

2 Tomlinson P B, The Botany of Mangroves (Cambridge, UK:

Cambridge University Press), 1986.

3 Hogarth P J, The Biology of Mangroves and Seagrasses

(Oxford, UK: Oxford University Press), 2007.

4 Manjunatha K R, Kumar T, Kundu N & Panigrahy S,

Discrimination of mangrove species and mudflat classes

using in situ hyperspectral data: a case study of Indian

Sundarbans, GIsci Remote Sens, 50 (4) (2013) 400-417.

5 Roughgarden J, Running S & Matson P, What does remote

sensing do for ecology, Ecology, 72 (6) (1991) 1918–1922.

6 Rama Rao N, Development of a crop‐specific spectral library

and discrimination of various agricultural crop varieties

using hyperspectral imagery, Int J Remote Sens, 29 (1)

(2008) 131-144.

7 Ustin S, Zarco Tejada P, Jacquemoud S & Asner G,

Remote sensing of environment: State the science and

new directions. In: Remote Sensing for Natural Resources

Management and Environmental Monitoring. Manual of

Remote Sensing, 3rd edn.

8 Kumar L, Schmidt K S, Dury S & Skidmore A K, Review of

hyperspectral remote sensing and vegetation science,

In: Hyperspectral remote sensing, edited by F Van Der Meer

(Dordrecht: Kluwer Academic Press), 2001.

9 Schmeller D S, European species and habitat monitoring:

where are we now? Biodivers Conserv, 17 (14) (2008)

3321–3326.

10 Van der Meer F, The effectiveness of spectral similarity

measures for the analysis of hyperspectral imagery,

Int J Appl Earth Obs, 8 (1) (2005) 3–17.

11 Clark R N, Swayze G A, Wise R A, Livo K E, Hoefen T M,

et al., USGS digital spectral library splib06a: U.S.

Geological Survey Digital Data Series 231, 2007.

12 Kokaly R F, King T V V & Livo K E, Airborne

Hyperspectral survey of Afghanistan 2007: flight line

planning and HyMAP data collection, Reston, VA, U.S.

Geological Survey, USGS Afghanistan Project Product No.

186, 2008.


756

13 Baldridge A M, Hook S J, Grove C I & Rivera G, The

ASTER spectral library version 2.0, Remote Sens Environ,

113 (4) (2009) 711–715.

14 Kruse F A & Perry S L, Improving multispectral mapping by

spectral modeling with hyperspectral signatures, J Appl

Remote Sens, 3 (1) (2009) pp. 033504.

15 Prasad K A, Gnanappazham L, Selvam V R,

Ramasubramanian R & Sekarkar C, Developing a spectral

library of mangrove species of Indian east coast using field

spectroscopy, Geocarto Int, 30 (5) (2015) 580-599.

16 Cochrane M A, Using vegetation reflectance variability

for species level classification of hyperspectral data,

Int J Remote Sens, 21 (10) (2000) 2175–2087.

17 Jensen R R, Hardin P J & Hardin A J, Classification of

urban tree species using hyperspectral imagery, Geocarto Int,

27 (5) (2012) 443-458.

18 Nidamanuri R R, Garg P K & Ghosh S K, Development of

an agricultural crops spectral library and classification of

crops at cultivar level using hyperspectral data, Precis Agric,

8 (2007) 173–185.

19 Wang L & Sousa W P, Distinguishing mangrove

species with laboratory measurements of hyperspectral

leaf reflectance, Int J Remote Sens, 30 (5) (2009)

1267-1281

20 Zomer R J, Trabucco A & Ustin S L, Building spectral

libraries for wetlands land cover classification and

hyperspectral remote sensing, J Environ Manage, 90 (7)

(2009) 2170–2177.

21 Broge N H & Leblanc E, Comparing prediction power and

stability of broadband and hyperspectral vegetation indices

for estimation of green leaf are index and canopy chlorophyll

density, Remote Sens Environ, 76 (2) (2000) 156–172.

22 Rama Rao N, Development of a crop‐specific spectral library

and discrimination of various agricultural crop varieties

using hyperspectral imagery, Int J Remote Sens Appl, 29 (1)

(2008) 131-144.

23 Korb A R, Dybwad P, Wadsworth W & Salisbury J W,

Portable FTIR spectrometer for field measurements of

radiance and emissivity, Appl Opt, 35 (1996) 1679–1692.

24 Selvam Y L, Gnanappazham M, Navamuniyammal,

Ravichandran K K & Karunagaran Y M, Atlas of Mangrove

Wetlands of India. Part I: Tamil Nadu, M. S. Swaminathan

Research Foundation, Chennai, 2002, pp. 100.

25 Vaiphasa C, Skidmore A K, Deboer W F & Vaiphasa T, A

hyperspectral band selector for plant species discrimination,

ISPRS J Photogramm Remote Sens, 62 (3) (2007) 225–235.

26 ASD, FieldSpec ® 3 User Manual, ASD Document 600540,

Rev J, 2010.

27 Gamon J A, Field C B, Bilger W, Bjorkman O, Fredeen A L,

et al., Remote sensing of the xanthophyll cycle and

chlorophyll fluorescence in sunflower leaves and canopies,

Oecologia, 85 (1) (1990) 1–7.

28 Sims D A & Gamon J A, Relationships between leaf pigment

content and spectral reflectance across a wide range of

species and developmental stages, Remote Sen Environ, 81

(2002) 337–354.

29 Vaiphasa C, Skidmore A K, Deboer W F & Vaiphasa T, A

hyperspectral band selector for plant species discrimination,

ISPRS J Photogramm Remote Sens, 62 (3) (2007) 225–235.

30 Wang L & Sousa W P, Distinguishing mangrove species

with laboratory measurements of hyperspectral leaf

reflectance, Int J Remote Sens Appl, 30 (5) (2009) 1267-

1281.

31 Song C, White B L & Heumann B W, Hyperspectral remote

sensing of salinity stress on red (Rhizophora mangle) and

white (Laguncularia racemosa) mangroves on Galapagos

Islands, Remote Sens Lett, 2 (3) (2011) 221-230.

32 Sahoo R N, Ray S S & Manjunath K R, Hyperspectral remote

sensing of agriculture, Cur Sci, 108 (5) (2015) 848-859.

33 Thenkabail P S, Enclona E A, Ashton M S & Meer B V D,

Accuracy assessments of Hyperspectral waveband

performance for vegetation analysis applications,

Remote Sen Environ, 91 (3-4) (2004) 354–376.

34 Schmidt K S & Skidmore A K, Spectral discrimination of

vegetation types in a coastal wetland, Remote Sen Environ,

85 (1) (2003) 92 – 108.

35 Rundquist D C, Field Techniques in Remote Sensing:

Learning by Doing, Geocarto Int, 16 (1) (2001) 85– 90.

36 Penndorf, Luminous and spectral reflectance as well as

colors of natural objects. U.S. Air Force Cambridge Research

Center, Bedford, Massachusetts, 1956.

37 Adam E & Mutanga O, Spectral discrimination of papyrus

vegetation (Cyperus papyrus L.) in swamp wetlands using

field spectrometry, ISPRS J Photogramm Remote Sens, 64

(2009) 612–620.

38 https://www.pngwing.com/en/free-png-bogao

development of spectral library of mangrove native species of...

Documents