A systems approach to analyze production options for wheat in India

Abstract

Systems tools such as simulation models, geographical information systems, databases and optimization techniques can be used to understand dynamic interactions among components of a production system. They can quantitatively analyze the agro-ecological properties of different land evaluation units in terms of biophysical and socio-economic factors, and their interactions. In this paper, we illustrate this for determining potentials, constraints and opportunities for further increase in productivity of wheat, a major food crop, in different regions of India. The whole country has been considered as a mega eco-region and its districts as sub eco-regions because most planning is done following these administrative boundaries.

In a large number of districts spread over the states of Punjab, Uttar Pradesh (D.P.), Bihar, Assam, Rajasthan, and Madhya Pradesh (M.P.), potential yields were 7 t ha¹ or more. Most districts in U.P. have a yield potential between 6 to 6.5 t ha¹. The potential yield was between 5 and 6 t ha¹ in middle latitudes and states of West Bengal and M.P. Economically optimal levels of N fertilizer application in irrigated environments were estimated for all locations based on current price ratios of N fertilizer and grain, native soil fertility, simulated crop response to N fertilizer and other costs related to transport, harvesting and market forces. A comparison of optimal and actual N applications showed that in Ludhiana district of Punjab, N application is more than the simulated optimal whereas in other districts it is at par or lower. The estimated yields corresponding to the profit maximizing antOunt of N apllication (henceforth refered as optimal economic yields) were generally 200 to 500 kg ha¹ lower than the potential yield irrespective of the location. The small difference between potential and optimal economic yield is due to distorted but favorable price ratios at present. In rainfed environments, optimal economic yields would be still lower.

At most locations, there was a large yield gap. At higher latitudes, the main wheat belt of India, yield gap of 2 t ha¹ was common even in well-irrigated regions. Almost 35-50% of the gap could be ascribed to delayed sowing, common in a number of districts. Factors such as limited and timely availability of irrigation and fertilizers, cropping pattern and access to credit and other services are some of the other principal causes of yield gaps.

It is concluded that crop growth simulation models together with databases of physical, biological and socio-economic attributes, geographical referencing and optimization techniques can help in setting up information systems to estimate crop production potentials, yield gaps, resource requirements for different agricultural strategies, assess potential environmental impacts, generate thematic maps and tables, and thus help in productivity related agro-ecological characterization.