Yu-Min Wang

A mixture neural methodology for computing rice consumptive water requirements in Fada N’Gourma Region, Eastern Burkina Faso

Crop consumptive water requirement (Crop-ET) is a key variable for developing management plans to optimize the efficiency of water use for crop production particularly in semiarid zone. In Burkina Faso, the unfavorable climatic conditions characterized by the low and unevenly distribution of rainfall have pushed water resources management to the forefront of the crop production issue. Crop-ET is extremely required in rainwater effective management for mitigating the impact of water deficit on the crops. Basically, Crop-ET determination involves reference evapotranspiration (ETo) and crop coefficient (Kc) which required complete climatic data and specific site crop information, respectively. ETo estimation with the recommended FAO56 Penman–Monteith (PM) equation is limited in Burkina Faso due to the numerous meteorological data required which are not always available in many production sites. In such circumstances, research to compute directly Crop-ET as an alternative to the two-step approach of calculating ETo and determining site specific Kc, seems desirable. Therefore, this study aims to evaluate the performance of a mixture principal component analysis neural network (PCANN) model for computing rice Crop-ET directly from temperatures data in Fada N’Gourma region located in Eastern Burkina Faso, Africa. From the statistical results, rice Crop-ET can be successfully computed by using PCANN methodology, when only temperatures data are available in this African semiarid environment. Thus, in poor data situation, Crop-ET direct computation can be rapidly addressed through PCANN model for agricultural water management in African semiarid regions.

Forecasting of Nonlinear Shoreline Variation Based on Aerial Survey Map by Neural Network Approach

Shoreline erosion problem may cause the possibility of losing land, and this must be considered particularly for a country which is composed of islands. In this study, six sandy nonlinear beaches located at Kenting National Park of Taiwan, were investigated in accordance with aerial survey maps taken in three different years. The images of spatial information were loaded into computerized graphical software, and the set fold function was used to make comparison for obtaining the tendency of sand line variation for the three different times. Based on the available dataset, a back propagation neural network model was then developed for forecasting the long term variation of sand line at each beach. The results showed that the sand line does have a phenomenon of rise and fall at some local regions, but the total area of each beach does not undergo significant changes and are within an acceptable range of error from a statistical standpoint. The neural network model used in this study might offer a new approach for solving this type of nonlinear problem, and the result obtained might provide a valuable reference for a relevant agency working in the area studied.

Using an integrated model to track the fate and transport of suspended solids and heavy metals in the tidal wetlands

Abstract An integrated two-dimensional depth-average numerical model was developed to simulate hydrodynamics and to track the fate and transport of contaminants in the Erh-Chung Flood Way wetland in northern Taiwan. The model was calibrated and verified with field data collected from Nov. 8, 2002 to April 13, 2003. The RMA2 and TOXIWASP models were configured for the wetland system to model mass transport, suspended sediment as well as heavy metals. Hydrodynamic results from the RMA2 model were used to develop mass transport for the TOXIWASP model in the wetland system. Model results mimic the field data for suspended sediment and heavy metals. Results of model sensitivity analyses show that the partition coefficient is a key parameter for the fate and transport of heavy metals in the wetland system.

Utilizing rainfall and alternate wetting and drying irrigation for high water productivity in irrigated lowland paddy rice in southern Taiwan

Abstract Taiwan’s average annual rainfall is high compared to other countries around the world; however, it is considered a country with great demand for water resources. Rainfall along with alternate wetting and drying irrigation is proposed to minimize water demand and maximize water productivity for lowland paddy rice cultivation in southern Taiwan. A field experiment was conducted to determine the most suitable ponded water depth for enhancing water saving in paddy rice irrigation. Different ponded water depths treatments (T2 cm, T3 cm, T4 cm and T5 cm) were applied weekly from transplanting to early heading using a complete randomized block design with four replications. The highest rainwater productivity (2.07 kg/m3) was achieved in T5 cm and the lowest in T2 cm (1.62 kg/m3). The highest total water productivity, (0.75 kg/m3) and irrigation water productivity (1.40 kg/m3) was achieved in T2 cm. The total amount of water saved in T4 cm, T3 cm and T2 cm was 20, 40, and 60%, respectively. Weekly application of T4 cm ponded water depth from transplanting to heading produced the lowest yield reduction (1.57%) and grain production loss (0.06 kg) having no significant impact on yield loss compared to T5 cm. Thus, we assert that the weekly application of T4 cm along with rainfall produced the best results for reducing lowland paddy rice irrigation water use and matching the required crop water.

Predictive accuracy of backpropagation neural network methodology in evapotranspiration forecasting in Dédougou region, western Burkina Faso

The present study evaluates the predictive accuracy of the feed forward backpropagation artificial neural network (BP) in evapotranspiration forecasting from temperature data basis in Dédougou region located in western Burkina Faso, sub-Saharan Africa. BP accuracy is compared to the conventional Blaney–Criddle (BCR) and Reference Model developed for Burkina Faso (RMBF) by referring to the FAO56 Penman–Monteith (PM) as the standard method. Statistically, the models’ accuracies were evaluated with the goodness-of-fit measures of root mean square error, mean absolute error and coefficient of determination between their estimated and PM observed values. From the statistical results, BP shows similar contour trends to PM, and performs better than the conventional methods in reference evapotranspiration (ET_ref) forecasting in the region. In poor data situation, BP based only on temperature data is much more preferred than the other alternative methods for ET_ref forecasting. Furthermore, it is noted that the BP network computing technique accuracy improves significantly with the addition of wind velocity into the network input set. Therefore, in the region, wind velocity is recommended to be incorporated into the BP model for high accuracy management purpose of irrigation water, which relies on accurate values of ET_ref.

Yield response, water productivity, and seasonal water production functions for maize under deficit irrigation water management in southern Taiwan

Abstract As the challenges toward increasing water for irrigation become more prevalent, knowledge of crop yield response to water can facilitate the development of irrigation strategies for improving agricultural productivity. Experiments were conducted to quantify maize yield response to soil moisture deficits, and assess the effects of deficit irrigation (DI) on water productivity (water and irrigation water use efficiency, WUE and IWUE). Five irrigation treatments were investigated: a full irrigation (I1) with a water application of 60 mm and four deficit treatments with application depths of 50 (I2), 40 (I3), 30 (I4), and 20 mm (I5). On average, the highest grain yield observed was 1008.41 g m−2 in I1, and water deficits resulted in significant (p < .05) reduction within range of 6 and 33%. This reduction was significantly correlated with a decline in grain number per ear, 1000-grain weight, ear number per plant, and number of grain per row. The highest correlation was found between grain yield and grain number per ear. The WUE and IWUE were within range of 1.52–2.25 kg m−3 and 1.64–4.53 kg m−3, respectively. High water productivity without significant reduction in yield (<13%) for I2 and I3 compared to the yield in I1 indicates that these water depths are viable practices to promote sustainable water development. Also, for assessing the benefits of irrigation practices in the region crop water production functions were established. Maize yield response to water stress was estimated as .92, suggesting the environmental conditions are conducive for implementing DI strategies.