Analysis of hydrological data for subsurface drainage design for an arid area in India

In the semi-arid to arid State of Rajasthan, 90% of the annual rainfall occurs as convective storms during the monsoon season. The suitability of a Markov chain analysis to model rainfall in this region was evaluated. A 5 day transition probability matrix accurately predicts sequences of wet and dry days based solely on the state of the preceding day. Both a daily model and an event-based model were used to describe rainfall pattern and distribution. Although the daily model is a traditional choice, the event-based model produced superior results. The event-based model describes the distribution of rainfall during a storm, the length of dry periods between rainfall events, and rainfall depths. The drainage coefficient derived from the normal value analysis, is 32.26 mm using the daily model, compared with 10.69 and 21.77 mm for the event-based models. The most cost-effective drainage coefficient is 10.69 mm derived from the 0.1 mm threshold event model. The Penman (1963) method best estimated ET[sub c] over both seasons as well as within each season. The Jensen-Haise method, when adjusted by a correction factor of 1.15 for ET[sub alfalfa], produced comparable estimates of ETc. The minimal climatic data required for the Jensen-Haise method makes it the most suitable evapotranspiration method for this area. A set of coefficients, ranging from 0.73 to 1.40, was developed to convert pan vaporation measurements to ET[sub JH]*. General crop coefficients for each development stage were determined from the generalized cropping pattern of the Chambal Command area. A water balance using effective wet year rainfall and evapotranspiration for the Kharif season was used to calculate the drainage requirement. The drainage requirement for the 178 ha Daglawada test plot is 749.2 and 1165.3 (x10³) m³, for return periods of 5 and 10 years, respectively. The leaching requirement of 0.0309, can be met with the Kharif season rainfall expected in wet years with return periods of 5 or more years, and normal years with return periods of 10 or more years.Applied Science, Faculty ofChemical and Biological Engineering, Department ofGraduat