Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Titre : Assessing irrigation performance by using remote sensing

Auteur : Bandara, K.M.P.S.

Université de soutenance : Wageningen Universiteit

Grade : PhD thesis 2006

Présentation
In Sri Lanka, most irrigation schemes irrigate rice and are located in the dry zone. The national annual average yield of rice is around 50 % of the genetic potential. In order to improve capabilities, productivity of paddy cultivation and objective data on actual field performance are needed. The improved satellite remote sensing can now provide information on a daily basis of biomass, soil moisture and ETact with spatial resolutions of 250 m to 1000 m. The main objective of this study was to develop and introduce a cost effective performance assessment program to manage irrigation systems using satellite remote sensing, and to assess whether the results are sufficiently accurate to support the managerial decisions at all levels.
The study area consists of the Uda­_Walawe and the Liyangastota irrigation schemes located in the South-east dry zone of Sri Lanka providing irrigation facilities for about 20000 ha of paddy. The Surface Energy Balance Algorithm for Land (SEBAL) was used for computing ETact over the cropped area with MODIS images. To assess operational performance, the temporal scale for each growth stage was considered for 10 day intervals. The indicators used were : relative evapotranspiration ETact / ETpot , delivery performance ratio Vc / Vc,int , depleted fraction ETact /( Vc + P ) and drainage ratio Vdr /( Vc + P ) (see Table 3.2).
During the wet and dry seasons, ETact fluctuated around 80 % of its potential value. During the wet season, the irrigation managers delivered more irrigation water than required . Due to this excess water delivery, as well as rainwater from surrounding highland areas which also flows into the drainage canal system, the drainage ratio Vdr/(Vc + P ) increased. In addition, by combining two indicators, the relative evapotranspiration ETact / ETpot and the depleted fraction ETact/(Vc + P ) a new water use matrix for irrigated crops was introduced in this thesis . Based on this matrix, each 10 day interval of the growth period is positioned in one of the four zones (see Fig.6.7). Thereby, the matrix describes how effectively the irrigation manager has delivered the irrigation water to reduce crop water stress.

Mots clés : rice / irrigation / remote sensing / irrigation scheduling / irrigation water / performance / crop yield / sri lanka

Présentation

Texte intégral