Journal of Pharmacognosy and Phytochemistry

Peer Reviewed Journal, Refereed Journal, Indexed Journal

E-ISSN: 2349-8234, P-ISSN: 2278-4136, NAAS Rating: 5.21
UGC Approved Journal. UGC Journal No.: 45051

This certificate confirms that "Dr. Kulkarni GB" has published manuscript titled "Cotton
vegetation condition monitoring using LSWI and NDVI".

Details of Published Article as follow:

Volume : 8
Issue : 3
Month : May-Jun
Year : 2019
Page Number : 1757-1762

Certificate No.: 8-2-371

Date: 01-05-2019

Yours Sincerely,

Akhil Gupta
Publisher
Journal of Pharmacognosy and Phytochemistry
www.phytojournal.com
Tel: +91-9711224068

Journal of Pharmacognosy and Phytochemistry

Email: [email protected] Website: www.phytojournal.com
 Journal of Pharmacognosy and Phytochemistry 2019; 8(3): 1757-1762

E-ISSN: 2278-4136
P-ISSN: 2349-8234
JPP 2019; 8(3): 1757-1762 Cotton vegetation condition monitoring using
Received: 18-03-2019
Accepted: 20-04-2019 LSWI and NDVI
Pritam O Bhutada
Asst. Sorghum Agronomist, Pritam O Bhutada, Dr. Kulkarni GB and RS Shinde
AICSIP, VNMKV, Parbhani,
Maharashtra, India
Abstract
This study aims to assess late-season agricultural drought in cotton growing Surendranagar district of
Dr. Kulkarni GB
Associate Professor,
Gujarat, India during Kharif cropping season 2015. Persistent abiotic (i.e. Moisture deficits and nutrient
Bagdia Arts, S.B. Lakhotia deficiency) during flowering and yield formation stage are stress referred to as late-season agricultural
Commerce & R. Bezonji Science drought. Satellite-based indices like the Normalized Difference Vegetation Index and Land Surface
College, Jalna, Maharashtra, Water Index from landsat-8 satellite data were analyzed. The analysis was carried out by comparing the
India satellite-derived indices with the previous normal years, and the assessments were made. The satellite-
based indices clearly brought out the stress that crop endured during September, while LSWI indicated
RS Shinde soil and crop water stress in early September. The result showed that NDVI and LSWI satellite based
Ph.D Scholar, Department of Vegetation indices which help to derived drought or stress affect cotton crop at critical growth stage very
Agronomy, Dr. BSKKV, well.
Dapoli, Maharashtra, India
Keywords: Cotton vegetation, LSWI and NDVI, Kharif

Introduction
Enhancing the efficiency of farm inputs without negatively impacting profitability or the
environment has been a primary focus of precision agriculture research for quite some time.
Plant production has to meet considerably mounting demands in the future. Expanding global
markets and the competition of food and non-food uses require further significant progress in
productivity levels. In India as well as globally, increased production will have to be achieved
on the same or decreasing area of arable land. Need to maintain or improve an increased
efficiency per unit area is required. At the same time, climatic changes may aggravate the
conditions of growth in less favorable locations. Thus, the scenario which agriculture facing is
further intensified the crop rotations with a limited number of high-yielding crops for the food
or raw materials market, under aggravated climatic conditions. Altogether, these developments
will result in a significant increase in problems caused by different stresses, which will
inevitably limit yield levels. To overcome on this problem, there is a need to spatial and timely
monitoring of crop season (F. Feldmann, D. V. Alford, c. Furk (eds.) 2009) [7].
Cotton (Gossypium hirsutum L.) belongs to the family Malvaceae. Cotton is a potentially
important commercial crop and plays a vital role in the textile industry in India. The history of
cotton production in India has spanned a period of more than seven decades. Gujarat is the
leading cotton producing state, near about 35% production and area of cotton in Gujarat. Area
of cotton in Gujarat 26.18 lakh ha and production is 104 lakh bales of 170 kg and cotton
Productivity 675 kg ha-1. In India, cotton is grown in about 122.35l lakh ha and production 377
lakh bales of 170 kg (ICAR-All India Coordinated Research Project on Cotton – Annual
Report (2017-18). It is obvious that growth and development of cotton has to face one or other
stress entities under rainfed. Cotton physiology portrays unique indeterminate growth habits
with longer crop duration which make cotton vulnerable to abiotic and biotic stress influences
from emergence to senescence. The adverse effects on the ongoing physiological processes
may affect yield projection trends leading to production lapses, inadequacies and may become
the focal point of attention. Cotton crop is one of the important cash crops and raw material for
textile industry. It is very sensitive for water stress condition. If drought or any stress occurs
during its critical growth stage i.e. maximum flowering to boll formation stage, it adversely
effects on yield of crop. Environmental stress conditions such as drought, heat, salinity, cold,
or pathogen infection can have a devastating impact on plant growth and yield under field
conditions. Water stress is important cause for adverse yield in agriculture. Drought is one of
Correspondence the important factors which effect on yield directly or indirectly such condition demand timely
Pritam O Bhutada and site specific crop monitoring. Satellite gives us chance for near-real time drought
Asst. Sorghum Agronomist,
AICSIP, VNMKV, Parbhani,
monitoring approach on crop condition using different vegetation indices.
Maharashtra, India
~ 1757 ~ 
Journal of Pharmacognosy and Phytochemistry
 
 
The objective of the study is to correlation of vegetation Patan and Kutch, South: Bhavnagar, East and South East,
indices derived by satellite data with different plant parameter West and South West: Rajkot District, Gujarat. Surendranagar
at critical growth of Cotton crop. district has found wide range of soil type such as alluvial soil,
sandy land of thin layer, medium type black and loamy soil
Study Area and less fertile land with thin layer. Medium type black soil is
Area is selected for study is Surendranagar, one of the found here and there. As the land is able to absorb moisture
important cotton producing district in Gujarat state of India. and sticky. Climate of Surendranagar is too hot in summer.
The chief agricultural product of Surendranagar district is Summer highest day temperature lies between 33 °C to 43 °C.
cotton. In fact Surendranagar district is one of the highest Average temperatures of January 22 °C, February 24 °C,
quality producers of cotton in world. First private cotton March 29 °C, April 32 °C, May 35 °C. Average annual
future trading exchange was established in Surendranagar. rainfall of area is 500-600 mm per annum. Major crops grown
Surendranagar is situated between 22° 43'N Latitude and 71° in area viz., Cotton, Green gram, Sesamum and Wheat.
43'E Longitude. Surendranagar is surrounded by North side:

Fig 1: Map of study area and cotton area masked

Methodology ’
Here we are using the Landsat image acquired from USGS
Earth Explorer. The data is in Geo-Tiff format with 16 bit Where:
radiometric resolution (ranges from 0-65535). Landsat 8 ρλ=TOA planetary reflectance, without correction for solar
Operational Land Imager (OLI) and Thermal Infrared Sensor angle. Note that ρλ’ does not contain a correction for the sun
(TIRS) images consist of nine spectral bands with a spatial angle.
resolution of 30 meters for Bands 1 to 7 and 9. The resolution Mρ=Band-specific multiplicative rescaling factor from the
for Band 8 (panchromatic) is 15 meters. In addition, it also metadata (Reflectance_Mult_Band_x where x is the band
has two Thermal IR bands with a spatial resolution of 100 m number)
(later re-sampled into 30 m). Before calculating the NDVI and Aρ=Band-specific additive rescaling factor from the metadata
LSWI the DN data must be converted to reflectance using the (Reflectance_Add_Band_x where x is the band number)
equations given in their websitehttp://landsatlook.usgs.gov. Q cal =Quantized and calibrated standard product pixel values
Here the IR and NIR bands are 4 and 5, respectively (DN)
(Francesco Pirotti et al., 2014) [8]. Reflectance with a correction for the sun angle is then:

Table 1: Bands and wavelength of Landsat 8 Operational Land ‘ ‘

Imager (OLI) and Thermal Infrared Sensor (TIRS)
cos sin
Bands λ(μm)
Band 1- Coastal aeroso l 0.433 - 0.453 Where:
Band 2-Blue 0.450 - 0.515 ρλ = TOA planetary reflectance
Band 3- Green 0.525 - 0.600 θSE = Local sun elevation angle. The scene center sun
Band 4- Red 0.630 - 0.680 elevation angle in degrees is provided in the metadata (Sun
Band 5- Near Infrared 0.845 - 0.885 Elevation).
Band 6- SWIR 1 1.560 -1.660 θSZ = Local solar zenith angle; θSZ = 90° – θSE
Band 7- SWIR 2 2.100 -2.300
Band 8- Pan 0.500 - 0.680 NDVI and LSWI
Band 9- Cirrus 1.360 - 1.390 Normalized Difference Vegetation Index (NDVI)
Band 10- TIRS 1 10.60 - 11.19 Normalized Difference Vegetation Index (NDVI) is a
Band 11-TIRS 2 11.50 - 12.51 measure of the amount and vigor of the vegetation at the
surface. The magnitude of NDVI is related to the level of
The following equation is used to convert DN values to TOA photosynthetic activity in the observed vegetation. In general,
reflectance for OLI image: higher values of NDVI indicate greater vigor and amounts of
~ 1758 ~ 
Journal of Pharmacognosy and Phytochemistry
 
 
vegetation. NDVI is a good indicator of green biomass, leaf information of drought also needs to be put in simpler terms
area index, and patterns of production. NDVI is computed (Chandrasekhar, K and Sesha Sai, 2015) [6]. Unlike point
using Red and NIR bands, which varies from -1 to +1. observations of ground data, satellite sensors provide direct
(Xiaojun Shea et al. 2014) [14]. spatial information on vegetation stress caused by drought
conditions. Satellite remote sensing technology is widely used

NDVI 
NIR  RED for monitoring crops and agricultural drought assessment.

NIR  RER Over the last 30 years, coarse resolution satellite sensors are
being used routinely to monitor vegetation and detect the
impact of moisture stress on vegetation. The NOAA AVHRR
Where, NDVI has been extensively used for drought/vegetation
ρNIR = reflectance in near infrared band monitoring, detection of drought and crop yield estimation
ρ RED = reflectance in red band (Batista et al., 1997; Moulin et al., 1998 [3] and Tucker et al.,
1985) [4].
Land Surface Water Index (LSWI) Monitoring the intensity and the density of the green
Studies have been conducted on VIs closely related to canopy vegetation, growth be measured using the reflection from the
water status. The Land Surface Water Index (LSWI) was red band and the infrared band. Green vegetation reflects
calculated as the normalized difference between the near- more energy in the near- infrared band than in the visible
infrared (NIR; 780−890 nm) and the shortwave-infrared range. It observe red band more for the photosynthesis
(SWIR; 1580 nm−1750 nm) wavelengths. process. From Fig.1 reflection of NDVI from cotton field
Land Surface Water Index (LSWI) is sensitive to change in initially shows minimum or less NDVI i.e. in -0.01 to 1 in
vegetation canopy water content month of June as just starting of land preparation or sowing of
cotton crop. As crop growth increased highest NDVI

LSWI 
NIR  SWIR observed during month of December after that again NDVI
started decline as crop harvesting started. The overall trend of
NIR  SWIR NDVI readings for each sensor generally increased with
increasing cotton crop growth stage and decline from crop
Where, maturity to crop harvesting.
ρNIR = reflectance in near infrared band Leaves reflect less in the near-infrared region when they are
ρSWIR = reflectance in short wave infrared band stressed, diseased or dead. Features like Clouds, water and
snow show better reflection in the visible range than the near-
Ground data collected infrared range, while the difference is almost zero for rock
From different sites of study area, we collected plant and soil and bare soil. We analyze the correlation between the plant
information in the month of September, 2015 at critical stage height, plant nitrogen% observed on dated 13th Sep, 2015,
of plant. NDVI and LSWI information obtained from vegetation
indices extracted from Landsat 8 datasets on same date. Most
Result and Discussion of studies shown that the thermal infrared is more sensitive to
No single index can fully describe the multi-scale, multi- acute water stress than is reflectance in visible, NIR, or SWIR
impact nature of drought in all its complexity. Hence, all wavelengths. However, the reflective portion of the spectrum
research efforts on drought monitoring and assessment are and VIs also respond to plant water status when it produces a
directed towards developing a composite index through change in canopy architecture, e.g., wilting or leaf rolling
appropriate blending techniques. In order to make it (Jackson and Ezra, 1985; Moran et al., 1989a) [10, 12]
acceptable by a wider section of drought management, the

Fig 2: Correlation OF NDVI with plant height and nitrogen uptake by plant

Plant height and Leaf nitrogen content correlates well to the and nitrogen uptake of cotton crop calculated by Kinjadal
NDVI data that also becomes more accurate as growth stage method, graph (Fig.2) shows linear correlation between these
increases., When we plot graph NDVI obtained from satellite parameter 0.54 and 0.55 R2 respectively.
data with ground data recorded on 13th Sept 2015, plant height
~ 1759 ~ 
Journal of Pharmacognosy and Phytochemistry
 
 
However, maximum NDVI of healthy cotton was 0.743, in leaf area and chlorophyll content, this will lead to a
whereas the maximum NDVI was 0.697 for infected cotton reduction of NDVI.
Most of the infected cotton died on average a month before
harvest (Mingquan Wu et al., 2016) [11], observed highest Land Surface Water Index
NDVI of healthy cotton crop field on 13th Sep 2015, was Due to the large fraction of surface water during the cropping
0.356 is very less. Cotton crop occurs continuously from June season, reflectance in the short-wavelength infrared (SWIR)
to September when the temperature was high and no rainfall range, which is sensitive to water, can be used to record the
at flowering and boll formation critical growth stage in the soil moisture status at the critical growth stage of cotton crop.
study area, because of this condition leaves dry and reduction

Fig 3: Correlation of LSWI with Soil Moisture and NDVI

The strong linear relationship between the indices is due to reached approximately 0.4.The decreasing trend of LSWI
the dominant impact of vegetation water content and the non- with NDVI was observed when crop at maturity stage and soil
flooded nature on the observed LSWI in the cotton fields. moisture decreases. From graph (Fig.3) R2=0.53 when plot
LSWI of the cotton showed increasing trend with NDVI. LSWI Vs soil moisture and R2= 0.90 plot LSWI Vs NDVI.
Even though LSWI is scattered widely around the linear fit, LSWI shows linear relation with soil moisture and NDVI of
LSWI increase linearly with NDVI until the NDVI of cotton crop.

Fig 4: Correlation of LSWI with Max. and Mini. Temperature

From graph (Fig.4) R2=0.56 when plot LSWI Vs Maximum LSWI decreases with increase in temperature as soil moisture
temperature and 0.45 plot LSWI Vs Minimum temperature. decrease.

~ 1760 ~ 
Journal of Pharmacognosy and Phytochemistry
 
 

Fig 5: Correlation OF NDVI and LSWI with plant stress area

Normalized difference vegetation index perform 3. Moulin SA, Bondeau A, Delecolle R. “Combining
exceptionally well when management goals require a agricultural crop models and satellite observations: from
quantitative means for tracking green biomass or leaf area field to regional scales, International Journal of Remote
index through the season or for detecting uneven patterns of Sensing. 1998; 19:1021-1036.
growth within a field (Jackson and Huete, 1991; Wiegand et 4. Tucker CJ, Townshend JRG, Goff TE. “African land
al., 1991) [9, 13]. LSWI was found to correspond well with the covers classification using satellite data”, Science, 227,
drought severities that were defined by the United States 369-375. Wang, Lingli., Qu, J. John., Hao, Xianjun and
Drought Monitor in previous studies. An LSWI-based drought Zhu, Qingping, 2008, “Sensitivity studies of the moisture
severity scheme is divided into four groups as extreme and effects on MODIS SWIR reflectance and vegetation
exceptional drought (LSWI ≤ −0.1), severe and moderate water studies”, International Journal of Remote Sensing,
drought (−0.1 < LSWI ≤ 0), abnormally dry (0 < LSWI ≤ 29, 7065-7075. Tucker, C. J. and Chowdhary, B. J., 1987,
0.1), and no drought (LSWI > 0.1) (Bajgain, R., 2015) [2]. “Satellite remote sensing of drought conditions”, Remote
During field survey we marked cotton crops which shows Sensing of Environment, 1985; 23:243-251.
stress condition due to lack of may be by nutrient, water, 5. Bausch WC, Diker K. Innovative remote sensing
insect pest etc because of these chlorophyll contain of leaves techniques to increase nitrogen use efficiency of corn,
decreases or leaf damaged. Such area marked depending on Communications in soil science and plant analysis, 2001;
condition of crop as : less than 1 is very poor,1- poor,2-good 32(7-8):1371-1390.
and 2.5-very good., However, when we plot graph of NDVI 6. Chandrasekar K, Sesha Sai. Monitoring of late-season
and LSWI obtained from satellite data from such field with agricultural drought in cotton-growing districts of Andhra
observed stress area marked during field survey, its (Fig.6) Pradesh state, India, using vegetation, water and soil
shows correlation R20.53 and 0.82 respectively. Graph shows moisture indices, M.V.R Nat Hazards. 2015; 75:1023.
that as moisture stress increases LSWI decreases same trend 7. Feldmann F, Alford DV, Furk C. (eds.) Crop plant
observed in case of NDVI. Result shows that LSWI and resistance to biotic and abiotic factors: current potential
NDVI were played good remote sensing indices to monitor and future demands. 3rd International Symposium on
vegetation condition in cotton crop. Plant Protection and Plant Health In Europe held at the
Julus Kühn - Institute, Berlin-Dahlem, Germany, 14-16
Conclusion may, 2009.
Remote sensing is one of best answer for the present problem 8. Francesco Pirottia, Maria Parragab A, Enrico Stuarob,
of drought in growing season of crop. NDVI and LSWI Marco Dubbinib, Andrea Masieroa, Maurizio
remote sensing indices which help to monitor crop and soil Ramanzinb. NDVI from landsat 8 vegetation indices to
condition and agriculture mid-season drought in many aspects study movement dynamics of capra ibex in mountain
of cotton crop. areas. the international archives of the photogrammetric,
remote sensing and spatial information sciences, volume
References xl-7, 2014, ISPRS technical Commission Vii
1. AICRPDA. District wise promising technologies for Symposium, 29 September – 2 October 2014, Istanbul,
rainfed cotton based production system in india a Turkey, 2014.
compendium of AICRPDA, CRIDA, AICCIP, SAUS, 9. Jackson RD, Huete AR. Interpreting vegetation indexes,
AN AICRPDA contribution. Compiled by K.P.R Vittal, Preventive Veterinary Medicine. 1991; 11(3-4):185-200.
T.P. Rajendran, G Ravindra, Chary, G.R Maruthi 10. Jackson RD, Ezra SE. Spectral response of cotton to
Sankart, Sri Jayays, Ramakrishna, J.S. Samra, Gurbachan suddenly induced water-stress, International Journal of
Singh. All India Coordinated Research Project, Central Remote Sensing. 1985; 6(1):177-185.
Research Institute For Dryland Agriculture, Hyderabad 11. Mingquan Wu, Chenghai Yang, Xiaoyu Song, Wesley
500 059, Andhra Pradesh, India, 2004, 01. Clint Hoffmann, Wenjiang Huang, Zheng Niu et
2. Bajgain R, Xiao X, Wagle P, Basara J, Zhou Y. al, Monitoring cotton root rot by synthetic Sentinel-2
Sensitivity analysis of vegetation indices to drought over NDVI time series using improved spatial and temporal
two tall grass prairie sites. ISPRS J Photogramm. Remote data fusion. Scientific Reports Article number, 2016,
Sens. 2015, 108:151-160. 2018, 8.
~ 1761 ~ 
Journal of Pharmacognosy and Phytochemistry
 
 
12. Moran MS, Jackson RD, Raymond H, Gay W, Slater PN.
Mapping surface-energy balance components by
combining landsat thematic mapper and ground-based
meteorological data. Remote sensing of environment,
1989; 30(1):77-87.
13. Wiegand CL, Richardson AJ, Escobar DE, Gerbermann
AH. Vegetation indexes in crop assessments. Remote
sensing of environment. 1991; 35:105-119.
14. Xiaojun Sheab, Lifu Zhanga, Muhammad Hasan Ali
Baigab, Yao Lia. Calculating vegetation index based on
the universal pattern decomposition method (VIUPD)
using landsat 8, IGARSS 2014, 4734-4737.
15. Annual Report. ICAR-All India Coordinated Research
Project on Cotton, 2017-18, 3-4.

~ 1762 ~