A GIS-based user-interface programme was developed to compute the geospatial Water ProductivityIndex (WPI) of a river-fed rice irrigation scheme in Northwest Selangor, Malaysia. The spatial analysisincludes irrigation blocks with sizes ranging from 20 to 300 ha. The amount of daily water use for eachirrigation block was determined using irrigation delivery model and stored in the database for both mainseason (August to December) and off season (February to May). After cut-off of the irrigation supply,a sub-module was used to compute the total water use including rainfall for each irrigation block. Therice yield data for both seasons were obtained from DOA (Department of Agriculture, Malaysia) of thescheme. Then, the Water Productivity Index (WPI) was computed for each irrigation block and spatialthematic map was also generated. ArcObjects and Visual Basic Application (VBA) programminglanguages were used to structure user-interface in the ArcGIS software. The WPI, expressed in termsof crop yield per unit amount of water used (irrigation and effective rainfall), ranged from 0.02 to 0.57kg/m3 in the main season and 0.02 to 0.40 in off season among irrigation blocks, respectively. Thedevelopment of the overall system and the procedure are illustrated using the data obtained from thestudy area. The approach could be used to depict the gaps between the existing and appropriate watermanagement practices. Suitable interventions could be made to fill the gaps and enhance water useefficiency at the field level and also help in saving irrigation water through remedial measures in theseason. The approach could be useful for irrigation managers to rectify and enhance decision-makingin both the management and operation of the next irrigation season.
In this research wok, three different techniques of change detection were used to detect changes in forest areas. One of the techniques used a local similarity measure approach to detect changes. This new approach of change detection technique, which used mutual information to measure the similarity between two multi-temporal images, was developed based on correspondence of the pixel values, rather than the difference in their intensity. Pixels suffering any changes will be maximally dissimilar. The study was conducted using multi-temporal SPOT 5 satellite images, with the resolution of 10 m x10 m on 5th August 2005 and 13th June 2007. The experimental results show that local mutual information provides more reliable results in detecting changes of the multitemporal images containing different lighting condition compared to the image differencing and NDVI technique, specifically in areas with less plant growth. In addition, it can also overcome the problem on selecting the threshold value. Besides, the findings of this study have also shown that band 3, which is sensitive to vegetation biomass, gave the best result in detecting area of changes compared to the others.
Precision agriculture with regard to crop science was introduced to apply only the required and optimal amount of fertiliser, which inspired the present study of nutrient prediction for oil palm using spectroradiometer with wavelengths ranging from 350 to 2500 nm. Partial least square (PLS) method was used to develop a statistical model to interpret spectral data for nutrient deficiency of nitrogen (N), phosphorus (P), potassium (K), magnesium (Mg), calcium (Ca) and boron (B) of oil palm. Prior to the development of the PLS model, pre-processing was conducted to ensure only the smooth and best signals were studied, which includes the multiplicative scatter correction (MSC), first and second derivatives and standard normal variate (SNV), Gaussian filter and Savitzky-Golay smoothing. The MSC technique was the optimal overall pre-treatment method for nutrients in this study, with highest prediction R2 of 0.91 for N and lowest RMSEP value of 0.00 for P.