Thumbnail Image

Anand Agricultural University, Anand

Anand Agricultural University (AAU) was established in 2004 at Anand with the support of the Government of Gujarat, Act No.(Guj 5 of 2004) dated April 29, 2004. Caved out of the erstwhile Gujarat Agricultural University (GAU), the dream institution of Sardar Vallabhbhai Patel and Dr. K. M. Munshi, the AAU was set up to provide support to the farming community in three facets namely education, research and extension activities in Agriculture, Horticulture Engineering, product Processing and Home Science. At present there seven Colleges, seventeen Research Centers and six Extension Education Institute working in nine districts of Gujarat namely Ahmedabad, Anand, Dahod, Kheda, Panchmahal, Vadodara, Mahisagar, Botad and Chhotaudepur AAU's activities have expanded to span newer commodity sectors such as soil health card, bio-diesel, medicinal plants apart from the mandatory ones like rice, maize, tobacco, vegetable crops, fruit crops, forage crops, animal breeding, nutrition and dairy products etc. the core of AAU's operating philosophy however, continues to create the partnership between the rural people and committed academic as the basic for sustainable rural development. In pursuing its various programmes AAU's overall mission is to promote sustainable growth and economic independence in rural society. AAU aims to do this through education, research and extension education. Thus, AAU works towards the empowerment of the farmers.

Browse

20031
Reset filters
Subject: Agricultural Meteorology × Author: AKULA, BABY ×
Reset filters

Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    ThesisItemOpen Access
    ESTIMATING WHEAT YIELDS IN GUJARAT USING WTGROWS AND INFOCROP MODELS
    (AAU, Anand, 2003) AKULA, BABY; Shekh, A. M.
    Crop simulation models are valuable tools to researchers to help them to understand the influence of climatic variables on crop productivity. The model estimated yields are handy to the agencies in government, trade and industry for planning about distribution, storage, processing, export or import of crop produce. Yield estimates by the models are also useful in taking timely policy decisions on fixing levy prices, because the estimates of the yield are available well in advance of the actual harvesting of the crop. Hence, a two-pronged approach was followed to estimate wheat yields in Gujarat, with the help of WTGROWS and InfoCrop simulation models. Initially both the models were calibrated and validated under Anand conditions through field experiment laid out in a strip plot design with three replications during rabi season of the years 2000 and 2001. Tliree dates of sowing were assigned as a main plot treatment with four irrigation regimes as sub plot treatments. Consistently higher yields were realised in case of the second date of sowing (15* Nov) during both the years although the yield differences were not statistically significant. Relatively more yields were realised in 2000 than those realised in 2001 and this was due to prevalence of favourable low temperatures during 50-90 DAS - a period that corresponded with anthesis to dough stage in conjunction with intermittent cold spells from 70-75 DAS corresponding with soft dough phase in the former year. In contrast to what was observed in case of yields in relation to the dates of sowing, yield data due to different irrigation treatments showed significant differences among them. Three irrigations gave significantly the lowest yield as compared with yields realised through any other irrigation treatment. The lowest yields realised in the treatment involving three irrigations were due to prevalence of moisture stress during tillering and flowering. Paradoxically, six irrigations despite not missing any important physiological stage, did not record significantly higher yield in comparison with yield in response to five irrigations. This was on account of the fact that, luxurious vegetative growth in the former case had caused lodging, as the prevailing wind speed was high. Different test criteria were followed to validate the performance of the models. Besides, error per cent was also calculated in all the different treatments to express the deviation in simulated values from those observed. Close scatter of simulated yield and total dry matter and respective measured values around the regression line and 1:1 line in case of both the models indicated good agreement between them. Both the models exhibited their robustness in predicting yields by explaining more than 90 per cent of variation in yield and total dry matter on an overall basis. However, there still remains some scope for improvement of the models in accounting for the loss due to lodging. The estimated RMSE for yield by WTGROWS was 318 kg ha-1, while that for yield by InfoCrop was 360 kg ha-1. Among the different dates of sowing, error per cent was relatively low in the treatments of the second date of sowing when compared with that for other dates. Both the models displayed decrease in error per cent with increase in irrigation levels. Underestimation of the simulated yield was more when the number of irrigations was less [three (I1) and four (I2)] when compared with that for more irrigations [five (I3) and six (I4)]. The underestimation was relatively more in case of InfoCrop, than that in case of WTGROWS. The performance of the models could be adjudged with the index of agreement (D), which was relatively high for WTGROWS (D= 0.97) than that for InfoCrop (D=0.95) in terms of yield. The models were also observed to perform in a similar way in terms of their response to the treatments in case of total dry matter, phenology and LAI also. The days to anthesis and maturity were simulated with less accuracy by both the models as compared to that of yield. Anthesis by WTGROWS explained more variance (R2=0.82) than that explained by InfoCrop (R2=0.75). The performance of these models in explaining the variance due to days to maturity was reverse of what was observed in case of anthesis. The highest and the .lowest ET were observed in case of the treatments of D2I4 and Dili, respectively. WTGROWS also showed similar pattern. Relatively higher proportion of MBE as compared to that of MAE during both the years in terms of ET as simulated by WTGROWS revealed under- prediction of ET by the model. Nonetheless, the error per cent did not cross the limit of -15 per cent during both the seasons except in case of Dill (-15.77%). Both the models expressed sensitivity to weather parameters viz., temperature, radiation and CO2 levels under both potential and stressed test conditions. But, the magnitude of change from the respective base yields in case of both the models was more to temperature under stressed conditions. However, the magnitude of response was more in case of WTGROWS than that in case of InfoCrop on overall basis except in case of radiation under stressed conditions where InfoCrop exhibited relatively more sensitivity. Linear response to TTVG, POTGWT, GNODMA, NSOILI, WLSTI was observed in case of both the models. The sensitivity was relatively more in case of WTGROWS than in case of InfoCrop. Moreover, InfoCrop exhibited linear response to RGRPOT and SLAVAR also. Statistical analysis of the historical actual wheat yield data of the state revealed that the average actual yield for the state as a whole was 2.5 t ha-1. Out of the ten districts selected to understand the temporal and spatial variability in wheat production levels and further to estimate yield gap through linking the model results with GIS, only Junagadh, Banaskantha and Bhavnagar exhibited significant positive linear trend at an average increase rate of 66, 31 and 25 kg ha-1 yr-1, respectively. Majority of the other districts failed to exhibit any discernible linear trend. However, Mehsana was found to be the second potential wheat producer of the state after Junagadh. The estimated average district potential yield by the models was 5.9 t ha-1 on overall bases. This is 2.36 times higher than the average actual state yield and is due to favourable thermal regimes as it was evident under Anand conditions where the estimated TTVG explained 87 per cent of variation in the potential yield and indicated significant linear positive trend. Similar reasoning holds good for higher potential yields in other districts. The attainable yields were estimated by imposing the management constraint of delayed sowing by twenty days from the optimum time (15thNov). The attainable wheat yields were found to decrease in all the districts irrespective of the agro climatic zone. The estimated attainable yield for the state as whole was 4.8 t ha-1 on the basis of the ten districts considered in the study. The average sowing yield gap between potential and attainable yield varied from 863 to 1205 kg ha-1 Reduction in yield due to delayed sowing was highest in the districts of Saurashtra which was followed in this respect by middle Gujarat, north Gujarat and south Saurashtra in sequence. The quantity of reduction in succession in these agro climatic zones was to the tune of 60, 59, 49 and 44 kg ha-1 per day delay in sowing, respectively.