Possibilities To Detect Supplementary Irrigation In Polder Areas With Remote Sensing

Possibilities To Detect Supplementary Irrigation In Polder Areas With Remote Sensing

Nick Madentzoglou, G.

Wim Bastiaanssen, W.G.M. (mentor)

Civil Engineering and Geosciences

Water Resources Management

MSc Geomatics

2008-06-30

Whenever a piece of land is reclaimed for some reasons it becomes a polder. Polder areas by definition lie below the mean sea level and constitute special environments with respect to the prevailed wet conditions. In Netherlands, where the biggest part of the country is below the mean sea level, polders are common features of topography. Three typical polders are Southern & Eastern Flevoland and the Northern polder. They have been established for more than five decades now and they are well-known for their agricultural character. Whenever we refer to agriculture or crops cultivation we always imply irrigation activities. Nevertheless, this implication is not fully true for polders, where many crops like sugar beets, maize, wheat etc. are never irrigated. On the other hand plants like flower bulbs and potatoes sometimes they are irrigated and sometimes not. It remains therefore a mystery, for water authorities which farmers actually irrigate and which not given that water consumption from irrigation is a crucial parameter for the management of water resources in the area. Therefore the main objective of the study is to locate farmers who irrigate (for a certain period) and to estimate this amount of water. Remote sensing is a powerful tool which has the potential to provide reliable and accurate water related information at a wide range of spatial (0.5-5000 m) and temporal resolutions (0.5-24 days). This information is essential for policy makers, consultants, water and irrigation managers as for public authorities. It becomes therefore challenging to investigate the potentiality of detecting irrigation with remote sensing. The growth of GIS has greatly enhanced the opportunity to integrate conventional and remote sensing data and the current research attempts to fully exploit it. In order to accomplish that, various data were selected ranging from remotely sensed e.g. ETact, ETpot, biomass production, land use to other water related data like seepage, precipitation, elevation and the water level maintained in polders’ canals. The research took place on 1995 which was a relatively dry year. From the land use map of the area potatoes, flower bulbs, orchards and other arable crops were selected for irrigation’s investigation. These crop types are evidently irrigated and also for June 1995 presented particularly different soil wetness levels. For the assessment of soil’s moisture state, ETdef was used as an indicator. The proposed method consists of two steps. First step involved the identification of wet and dry plots based only on satellite data (ETdef). Certain ETdef thresholds were drawn in order to facilitate that distinction. Following dry plots were discarded and the analysis was carried with wet ones. Second step involved the identification of soil’s wetness sources. The potential sources which allow a plot to appear wet are: precipitation, seepage, soil moisture (depending on soil type), the water level maintained in canals and finally irrigation. The impact of each source was assessed for every individual (wet) plot and in cases where this was negligible wetness was attributed to irrigation, otherwise not. This process resulted in a final irrigation map, consisting of 9 distinct classes ranging from potentially irrigated to non-irrigated plots. Moreover the volume of water which was used for irrigation was calculated for each plot separately. Findings revealed that on the 25th of June 37.2% of potatoes, 51.3% of flower bulbs, 44% of orchards and 37.1% of other arable crops were irrigated by farmers. The total amount of water used for irrigation was approximately 184229 m3 day-1 ha-1. Unfortunately validation of the results was possible only qualitatively since no relevant field data were available for comparison. Interviews with local water managers revealed that final irrigation maps depict reality accurately and describe the situation very close to actual conditions. Overall the objectives of the study were accomplished and from now on polders’ irrigation is no longer a mystery.

Remote Sensing
Irrigation

http://resolver.tudelft.nl/uuid:5242692a-7d47-4e72-93b7-8639c4cc2696

Student theses

master thesis

(c) 2008 Nick Madentzoglou, G.