The University of Southampton
University of Southampton Institutional Repository

Classification of vegetation type in Iraq using satellite-based phenological parameters

Classification of vegetation type in Iraq using satellite-based phenological parameters
Classification of vegetation type in Iraq using satellite-based phenological parameters
Primary information of great importance to various grand challenges such as sustainable agricultural intensification, food insecurity, and climate change impacts, can be obtained indirectly from land cover monitoring. However, in arid-to-semiarid regions, such as Iraq, accurate discrimination of different vegetation types is challenging due to their similar spectral responses. Moreover, Iraq has been subjected to major disturbances, both natural and anthropogenic which have affected the distribution of land cover types through space and time. Reliable information about croplands and natural vegetation in such regions is generally scarce. This research aimed to develop a phenology-based classification approach using support vector machines for the assessment of space-time distribution of the dominant vegetation land cover (VLC) types in Iraq, particularly croplands, from 2002 to 2012. Thirteen successive years of 8-day composites of MODISNDVI data at a spatial resolution of 250 m were employed to estimate 11 phenological parameters. The classification methodology was assessed using reference samples taken from fine spatial resolution imagery and independent testing data obtained through fieldwork. Overall accuracies were generally >85%, with relatively high Kappa coefficients (>0.86) across the classified land cover types. The predicted cropland class area and the Global MODIS land cover product were compared with ground statistical data at the governorate level, revealing a significantly larger coefficient of determination for the present phenology-based approach R2 = 0.70 against R2 = 0.33 for MODIS, p< 0.05. The resulting maps delimit for the first time, at a fine spatial resolution, the spatial and interannual variability in the dominant VLC classes across Iraq.
Classification, Iraq, monitoring, phenology, time-series
1939-1404
414-424
Qader, Sarchil
b1afb647-aeff-4bb8-84f2-56865c4eb9e4
Dash, Jadunandan
51468afb-3d56-4d3a-aace-736b63e9fac8
Atkinson, Peter M.
96e96579-56fe-424d-a21c-17b6eed13b0b
Rodriguez Galiano, Victor
44144f72-19cd-433e-be40-36a054d8fbf3
Qader, Sarchil
b1afb647-aeff-4bb8-84f2-56865c4eb9e4
Dash, Jadunandan
51468afb-3d56-4d3a-aace-736b63e9fac8
Atkinson, Peter M.
96e96579-56fe-424d-a21c-17b6eed13b0b
Rodriguez Galiano, Victor
44144f72-19cd-433e-be40-36a054d8fbf3

Qader, Sarchil, Dash, Jadunandan, Atkinson, Peter M. and Rodriguez Galiano, Victor (2016) Classification of vegetation type in Iraq using satellite-based phenological parameters. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 9 (1), 414-424. (doi:10.1109/JSTARS.2015.2508639).

Record type: Article

Abstract

Primary information of great importance to various grand challenges such as sustainable agricultural intensification, food insecurity, and climate change impacts, can be obtained indirectly from land cover monitoring. However, in arid-to-semiarid regions, such as Iraq, accurate discrimination of different vegetation types is challenging due to their similar spectral responses. Moreover, Iraq has been subjected to major disturbances, both natural and anthropogenic which have affected the distribution of land cover types through space and time. Reliable information about croplands and natural vegetation in such regions is generally scarce. This research aimed to develop a phenology-based classification approach using support vector machines for the assessment of space-time distribution of the dominant vegetation land cover (VLC) types in Iraq, particularly croplands, from 2002 to 2012. Thirteen successive years of 8-day composites of MODISNDVI data at a spatial resolution of 250 m were employed to estimate 11 phenological parameters. The classification methodology was assessed using reference samples taken from fine spatial resolution imagery and independent testing data obtained through fieldwork. Overall accuracies were generally >85%, with relatively high Kappa coefficients (>0.86) across the classified land cover types. The predicted cropland class area and the Global MODIS land cover product were compared with ground statistical data at the governorate level, revealing a significantly larger coefficient of determination for the present phenology-based approach R2 = 0.70 against R2 = 0.33 for MODIS, p< 0.05. The resulting maps delimit for the first time, at a fine spatial resolution, the spatial and interannual variability in the dominant VLC classes across Iraq.

Full text not available from this repository.

More information

Accepted/In Press date: 2 December 2015
e-pub ahead of print date: 5 January 2016
Published date: 28 January 2016
Keywords: Classification, Iraq, monitoring, phenology, time-series

Identifiers

Local EPrints ID: 412164
URI: https://eprints.soton.ac.uk/id/eprint/412164
ISSN: 1939-1404
PURE UUID: 24ab9198-4559-40eb-9436-08ae7f46861c
ORCID for Jadunandan Dash: ORCID iD orcid.org/0000-0002-5444-2109
ORCID for Peter M. Atkinson: ORCID iD orcid.org/0000-0002-5489-6880

Catalogue record

Date deposited: 13 Jul 2017 16:31
Last modified: 10 Dec 2019 01:55

Export record

Altmetrics

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×