Abstract
Field studies for jute production system were conducted for two jute growing seasons (2006 and 2007) at Central Research Station of Bangladesh Jute Research Institute, Manikgonj to study the growth patterns for different plant components at the development stages of two varieties of jute. There was no significant difference in the growth patterns in leaf, root, bark and stick dry matter and photosynthetic leaf area of the two varieties of jute. Emergy analysis of jute production showed that emergy sustainability indicators of jute production for 10 years (1997-2006) were very good. But the returns on investments were very low which indicated that the price paid for emergy of the produce was extremely poor. It revealed that the jute production is emergetically efficient and it does not impose any extra load to the environment, but the emergy value of the produce is not profitable. The life cycle assessments of jute production systems in Bangladesh were done for two distinct retting processes. Improved retting process consumed less input but produces almost same output and emitted less to water and air compared to the conventional one and the jute production in Bangladesh is environmentally sound. A series of artificial neural network models were developed to predict the growth patterns of the different organs of the jute plant and the photosynthetic leaf area growth patterns. The models consisted of four-layered networks and a large number of neurons. Two sets of experiments conducted in 2006 were used for training the models and another two sets of experiments conducted in 2007 were used for testing of the models. Predicted results showed an excellent agreement with the observed results. A system dynamics model FIBGROW was developed to predict the impacts of climate change on jute production in Bangladesh. The model was validated comparing the simulated values with the observed data. The agreements between the simulated and observed values were very good. Sensitivity analyses of the important parameters showed that initial light use efficiency of single leaf, maximum rate of photosynthesis of single leaf, extinction coefficient and initial leaf area value were found to be sensitive. Crop management strategies of plant density, sowing date, and their interactions were simulated and the predicted results agreed well with the reported values. Sixteen treatment combinations of temperature increment (0°C, +1°C, +2°C, and +3°C) and solar radiation change (0%, +5%, +10%, and -5%) and twenty treatment combinations of CO2 concentration (390, 440, 490, 540 and 640 ppm) with the increase in average temperature (0°C, +1°C, +2°C and +3°C) were considered for fibre yield simulation. Increased fibre yield was simulated for the higher solar radiation and elevated CO2 concentrations for 1°C increase in temperature but fibre yield decreased for the temperature rise of 2°C and 3°C. Fibre yields were simulated decadewise climatic variabilities up to the 2100 AD using the climate change prediction scenarios of IPCC and different GCMs. Fibre yields of jute are more or less same up to 2030. Simulated fibre yields for the predicted climate change scenarios showed a clear tendency of decreasing yield after 2030, which eventually falls down to 57.54% at the end of 2100 AD. Finally, FIBGROW model is an excellent tool for predicting potential yields, management strategies and climate change impacts on jute production system.