Environ Monit Assess (2009) 159:281���292 DOI 10.1007/s10661-008-0629-3 Assessment of temporal variations of water quality in inland water bodies using atmospheric corrected satellite remotely sensed image data Diofantos G. Hadjimitsis �� Chris Clayton Received: 11 January 2008 / Accepted: 5 November 2008 / Published online: 6 December 2008 �� Springer Science + Business Media B.V. 2008 Abstract Although there have been many studies conducted on the use of satellite remote sensing for water quality monitoring and assessment in in- land water bodies, relatively few studies have con- sidered the problem of atmospheric intervention of the satellite signal. The problem is especially significant when using time series multi-spectral satellite data to monitor water quality surveillance in inland waters such as reservoirs, lakes, and dams because atmospheric effects constitute the majority of the at-satellite reflectance over water. For the assessment of temporal variations of wa- ter quality, the use of multi-date satellite images is required so atmospheric corrected image data must be determined. The aim of this study is to provide a simple way of monitoring and assessing temporal variations of water quality in a set of inland water bodies using an earth observation- D. G. Hadjimitsis (B) Department of Civil Engineering and Geomatics, Faculty of Engineering and Technology, Cyprus University of Technology, P.O. Box 50329, 3036, Lemesos, Cyprus e-mail: d.hadjimitsis@cut.ac.cy, fantos@cytanet.com.cy URL: http://www.cut.ac.cy/ce/ C. Clayton School of Civil Engineering and the Environment, University of Southampton, Highfield, Southampton, SO17 1BJ, UK e-mail: c.clayton@soton.ac.uk based approach. The proposed methodology is based on the development of an image-based al- gorithm which consists of a selection of sampling area on the image (outlet), application of mask- ing and convolution image processing filter, and application of the darkest pixel atmospheric cor- rection. The proposed method has been applied in two different geographical areas, in UK and Cyprus. Mainly, the method has been applied to a series of eight archived Landsat-5 TM images acquired from March 1985 up to November 1985 of the Lower Thames Valley area in the West London (UK) consisting of large water treatment reservoirs. Finally, the method is further tested to the Kourris Dam in Cyprus. It has been found that atmospheric correction is essential in water quality assessment studies using satellite remotely sensed imagery since it improves significantly the water reflectance enabling effective water quality assessment to be made. Keywords Outlet �� Temporal variations �� Water quality �� Atmospheric correction �� Remote sensing Introduction It has been shown by several other researchers, who studied the actual relationship between water properties (i.e., water quality) and satellite

282 Environ Monit Assess (2009) 159:281���292 data for several types of water bodies and geo- graphical extensions, that satellite remote sens- ing techniques show more important advantages than traditional sampling (Dekker et al. 1995 Pulliainen et al. 2001 Chen et al. 2007 Pozdnyakov et al. 2005 Hadjimitsis et al. 2006 Ormeci et al. 2008 Weiqi et al. 2008). Firstly, the continuous geographical coverage of satellite imageries pro- vides continuous water quality information about the whole water body. Secondly, remote sensing allows us to obtain information about inaccessible places. Finally, historical satellite images provide estimation of historical water quality and offer an excellent way to monitoring the water quality temporal evolution as well. Weiqi et al. (2008) re- ported that, with the development of remote sens- ing techniques, water quality monitoring based on remote sensing methods has become accessible and very efficient. Although several satellite remote sensing sys- tems have been used for water quality assessment and monitoring, the relatively low cost, temporal coverage, spatial resolution, and data availability of the Landsat system make it particularly useful for water quality assessment and monitoring of in- land water bodies (Hadjimitsis et al. 2000a, 2006). Several studies have demonstrated a strong rela- tionship between Landsat-5/7 Thematic Mapper (TM) or Enhanced Thematic Mapper (ETM+) data and ground observations of water qual- ity parameters such as chlorophyll-a, suspended solids, turbidity, etc. (e.g., Lillesand et al. 1983 Carpenter and Carpenter 1983 Lathrop and Lillesand 1986 Lathrop 1992 Baban 1993 Arenz et al. 1996 Cox et al. 1998 Giardino et al. 2001 Kloiber et al. 2002a, b Dewidar and Khedr 2001 Wang et al. 2004 Vincent et al. 2004 Hadjimitsis et al. 2006). Most of the published works in which Landsat MSS, TM, Spot HRV, or other image data have been used for monitoring inland water quality studies comprise the use of sampling mea- surements in combination with the digital imagery either using radiance or reflectance (Harris et al. 1976 Verdin 1985 Lathrop and Lillesand 1986 Gitelson et al. 1993 Mayo et al. 1995 Arenz et al. 1996 H��rm�� et al. 2001 Kloiber et al. 2002a, b Dekker et al. 2001 Zhang et al. 2002, 2003 Chen et al. 2007). Many researchers have investigated the temporal changes of water quality in inland water bodies using ground data (for example, Nellis et al. 1998). The use of satellite remote sensing can only replace the ground campaigns after continuous testing and calibration of the proposed image-based methods and after consid- ering the atmospheric effects to the at-satellite receiving signal. Many inland water quality stud- ies have used non-calibrated and non-atmospheric corrected data, producing results that are nei- ther comparable temporally nor spatially (Verdin 1985 Chen et al. 2007). Optical remote sensing data are affected by the atmosphere. These ef- fects can be removed using suitable atmospheric corrections and bi-directional reflectance models (Zhang et al. 2003 Hadjimitsis and Clayton 2008). To overcome these limitations, the authors pro- vide a new method for assessing temporal varia- tions on the sampling point (outlet) of each inland water body such as water treatment reservoir or water dam. Materials and methodology Study areas The proposed method has been applied into two different geographical areas, the Lower Thames Valley in vicinity of Heathrow Airport in the UK and the Kourris Dam located in the Limassol District area in Cyprus. Lower Thames Valley, West London (UK) The first study area is located to the south and the west of London Heathrow Airport in the UK. It includes many inland water bodies such as reservoirs, rivers, lakes, and ponds. Emphasis has been given to the larger reservoirs in the Lower Thames Valley (see Fig. 1). The Lower Thames Valley Reservoirs are characterized as eutrophic and are used for a number of purposes such as for storage of water, as the first stage of potable water treatment, and for recreational purposes. Despite the fact that cloud cover is a major problem in this area (Hadjimitsis et al. 2004), it has been found that satellite remote sensing and especially Landsat TM imagery can be a useful tool for mon- itoring water quality in such reservoirs as well to assist the water resources managers to locate new