Search

Closed-loop control for precision irrigation represents an effective method to provide water savings in this resource-intensive activity. Typically, implementation of these systems uses an on-off control approach with soil moisture as the feedback variable. In these cases, no process modeling is required due to the simplicity of the implemented control algorithm. However, the amount of water consumed by irrigation globally presents an interesting application area for the control discipline. Therefore, to obtain greater water savings and to improve crop productivity through control theory, an irrigation model is required that integrates the three main elements in the irrigation process: soil, crop, and weather. This article proposes a hybrid automata model for a closed-loop irrigation system applied to a grass-covered area. The main measured variables used to represent the process dynamics are soil moisture, vegetation index, and evapotranspiration. The hybrid automata approach allows a complex system to be modeled as a finite state machine in which there is a specific linear model for each state. When certain conditions are met, the system transitions from one state to another, and the new behavior is represented by a different linear model. To develop the proposed model, experimental data were obtained from an irrigation process, the data were analyzed to produce a model using the hybrid automata approach, and finally the model was validated with a new set of measured data. [ABSTRACT FROM AUTHOR]

This study assessed the impact of potential future climate change on the inflow to an agricultural reservoir and its downstream streamflow from reservoir operation for paddy irrigation using SLURP (semi-distributed land use-based runoff process), a physically based hydrological model. For a 366.5 km² watershed that includes two agricultural reservoirs located in the upper middle part of South Korea, the SLURP model was calibrated using six years (1998-2003) of daily streamfow records and validated using three years (2004-2006) of streamflow data, before the future analysis. The CCCma CGCM2 data of the SRES (special report on emissions scenarios) A2 and B2 scenarios of the IPCC (Intergovernmental Panel on Climate Change) was adopted for the future climate condition, and then the change factor method was used to downscale the future weather data through bias-correction. In addition, the future land covers were predicted with a modified CA (cellular automata)-Markov technique using a time series of Landsat land cover data, and the future vegetation information was predicted by linear regression between the monthly NDVI (normalized difference vegetation index) from NOAA AVHRR (advanced very high resolution radiometer) images and the monthly mean temperature data. The results for the future impact on reservoir storage showed that it would not have recovered to 100% by the end of the year (December) because of a decrease in precipitation and an increase in evaporation from July to September. Therefore, this reduced reservoir storage could bring a spring drought that causes an insufficient guarantee of water for rice paddy transplanting at the end of May. The future impact on the monthly streamflow of the watershed showed that the future streamflow for the three months of April, May, and June would increase because of a future increase in precipitation, and that the future streamflow for the next three months (July, August, and September) would decrease because of a future decrease in precipitation and increase in temperature. Considering the future decrease in the summer and autumn streamflows, the reservoir operation has to be more conservative when supplying water for paddy irrigation during this period and in the decision making process for reservoir releases during storm events. [ABSTRACT FROM AUTHOR]