Reuse of upland drainage waters has become an acceptable and common practice among many farmers who have no access to good-quality irrigation water, and suffer long periods of droughts in and and semi-arid regions. This study was carried out in a farmer's cotton field of 0.27 ha, located in the Eastern Mediterranean Coastal Region of Turkey, at 2.1 m mean sea level. The area presently lacking irrigation water has a typical Mediterranean climate with dry and hot summers, and cool and rainy winters. The farmers in the area use low-quality irrigation water, diverted from drainage channels, carrying irrigation return flows of upland fields. The objective of the work was to assess what effect the existing practice of irrigation can have on soil salinity using both conventional statistics and geostatistical techniques. Eighty one soil samples were collected from 0 to 30 cm depth in 1999 and 2000 along five parallel transects established in North/South direction of the experimental field. Two samplings were done each year, before and after irrigation seasons, in early June and late September, respectively. Simple-mean ECe-comparison tests of the data revealed that soil salinity had decreased from 4.8 to 3.0 dS m(-1) under farmer's irrigation practice over the two irrigation seasons. There was always a decreasing trend in soil salinity from the beginning until completion of the irrigation seasons, in both years 1999 and 2000. Kriged contour maps, drawn based on spatial variance structure of the data, revealed that saline areas (ECe greater than or equal to 4 dS m(-1)) of the field decreased from 51 to 20%, confirming the trend demonstrated with conventional statistics. The results, although limited only to 2 years work, suggest that the risk of increasing soil salinity is nearly nil would farmers cultivate and irrigate downstream low lands where irrigation schemes can not reach, rather than using these areas only for rain-fed farming. (C) 2002 Elsevier Science B.V. All rights reserved.