Farmers in Madhya Pradesh grow a soybean crop during the monsoon season, and then grow a wheat crop during the dry season using moisture stored in the deep vertosol soil, supplemented with some irrigation from groundwater. While the farmers do use fertilizer, this is restricted to modest rates of nitrogen and phosphorus application supplemented with farm yard manure (FYM). Application rates of nitrogen and phosphorus are less than the rate of removal, and there has been little consideration given to other nutrients. The farmers obtain good wheat yields (though lower than the national average), but typical soybean yields are less than expected in this environment.

Nutrient omission trials were conducted for the wheat crop on farmers fields in Geelakhedi (Rajgarh district), Mugaliahat (Bhopal district) and Rangai (Vidisha district) villages. Two replicated experiments were established in each village. In an omission trial, all but one nutrient is applied to the crop. If the omission of a nutrient results in a yield lower than the control plot (where all nutrients are applied), then the nutrient omitted is recognised as deficient in the soil. In addition to dry-matter and grain yield measurements, plant tissues samples and soil samples were also taken for chemical analysis. This approach identified phosphorus, sulfur and zinc as deficient, but showed that adequate levels of iron, manganese, copper, boron and molybdenum were present. Response to nitrogen application was not tested, as this nutrient is recognised by farmers as being deficient, and is routinely applied as fertilizer. The response to phosphorus on several of the field sites, despite regular applications of phosphatic fertilizer, indicates that the rate of application used has only been sufficient to maintain production, and that no reservoir of phosphorus has accumulated. A reduced program of nutrient omission trials has been established for the 2005-2006 year soybean crop.

A farmer's practice plot was also included to increase the applicability of the trial as an extension tool. The farmers practice plots produced a lower yield than the control in all six trials, effectively demonstrating that the farmer was loosing yield because of poor crop nutrition. The research trials attracted considerable farmer attention, with many farmers making "casual" visits to the site. In addition, several formal village meetings were held to discuss the field trials and the direction of future research. Future interactions with the farming community, and in particular discussion of strategies for overcoming yield limitations, will be fostered by the appointment of two BAIF community workers, and through the appointment of a social scientist at the Indian Institute of Soil Science.

Neal Menzies, Pax Blamey and Ram Dalal visited India in August 2004 (before the project document was signed) to make a preliminary assessment of the soybean crop. This visit was timed to permit the researchers to observe the soybean crop in its late vegetative / early reproductive stage. It was considered that at this time, foliar symptoms of nutrient deficiencies would be most apparent, and that a visit at this time would help the research team to understand the agronomic practices used by Indian farmers for soybean culture. The timing of this visit was an ideal opportunity to assess the impact on the soybean crop of waterlogging; the visit following several weeks of heavy rain. In low-lying areas of the field, where the soil surface was covered with standing water, the soybean crop was stunted and profoundly nitrogen deficient. However, over the majority of the field area, there was no standing water and the crop appeared healthy (plants were green and turgid). Furthermore, while there was some insect damage, and in particular damage from the stem girdling beetle, this was not thought to be sufficient to cause a substantive yield loss. Several hypotheses to account for the poor soybean yield, despite adequate vegetative growth were developed for testing. The potential limitations to yield considered included, low photosynthetically active radiation levels during the overcast days of the monsoon, and that an early end to the monsoon resulted in water stress during pod-filling. This latter effect could be accentuated by poor exploitation of the soil profile for water, as much of the plants root system in the deeper part of the soil profile may have been damaged by anoxic conditions during waterlogging. The use of a broad bed and furrow (BBF) system would overcome the problems posed by complete inundation of the soil. It may also increase the depth of "healthy" roots during the monsoon, and thus may alleviate some of the problems of rapid onset of drought conditions at the end of the monsoon.

In addition to field experimentation, crop simulation modelling will be used to explore limitations to soybean yield. To provide appropriate crop physiology parameters for this modelling, an assessment of soybean phenology and biomass accumulation was planned, and has been implemented for the 2005-2006 year.

In Australia, no field trials were initiated because of the ongoing drought. In view of this limitation to the progress of the project, the research assistant position at the Department of Natural Resources and Mines was not filled.

Nutrient omission trials were conducted for the wheat crop on farmers fields in Geelakhedi (Rajgarh district), Mugaliahat (Bhopal district) and Rangai (Vidisha district) villages (two replicated trials in each village), followed by a smaller number of replicated trials in the soybean crop. For both crops, P, S and Zn were identified as deficient. Fertilizers containing N (not tested, but known to be deficient), and P are routinely used; most commonly di-ammonium phosphate (DAP). Sulfur fertilizer is not commonly applied, and this deficiency may be a widespread limitation to crop production. For the soybean crop, the use of superphosphate fertilizer, which contains both P and S, would be ideal. Unfortunately, the superphosphate available in the marketplace is not of a consistently high quality, and farmers are reluctant to purchase it.

Integrated nutrient management (the combination of nutrient supply from organic matter and inorganic fertilizers) trials were conducted in Raigarh and Bhopal districts (two replicated trials in each district). Treatment regimes were developed to evaluate inorganic fertilization, organic fertilization, and a combination of inorganic fertilizer and organic fertilization for the soybean/wheat cropping system. In the soybean season, the treatments used included inorganic N, P, K, S and Zn at recommended rates, 5t FYM + 50% of the recommended rate of inorganic fertilizer, 5t farm yard manure (FYM) + 50% inorganic fertilizer + rhizobium inoculation, and an organic matter treatment of 5t FYM. Other treatments evaluated the effectiveness of phosphorus solubilising bacteria, and Azotobacteria (a free living N fixer). A farmers' practice treatment was also included. Of the treatments used, the highest soybean grain yields were obtained in the 100% inorganic fertilizer (2.1 t/ha), 50% inorganic + 5t FYM (2.2 t/ha), and 50% inorganic + 5t FYM + rhizobium inoculation (2.3 t/ha); these treatments were not significantly different to each other, but yielded significantly more than the other treatments tested (1.8 to 1.9 t/ha). In the wheat season, these same treatments also produced the best grain yields of 4.8 to 5.0 t/ha (the 100% inorganic fertilizer treatment again receiving the recommended rates of N, P, K, S and Zn, and the integrated nutrient management treatments receiving 75% of the recommended rate of inorganic fertilizer). The organic fertilization treatments, which received 8t FYM prior to planting the wheat, produced significantly lower yields (around 4t/ha). These top performing treatments were also the most attractive from an economic perspective.

The inclusion of a farmer's practice plot within the trials was a useful extension tool; particularly as the farmers hosting the trials are recognised as the leading farmers in the district. The farmers in the villages hosting the experiments viewed the result obtained with the 50% inorganic + 5t FYM + rhizobium treatment as particularly attractive to them, and numerous farmers state the intention of using this fertilization regime in the next soybean season. The key fertilization treatments (100% inorganic, 50% inorganic + 5t FYM + rhizobium, and the farmers' practice) will be repeated in a large number of "baby trials" to confirm the result in a wide range of settings, and as a means of extending the outcomes of the integrated nutrient management trial to a large number of farmers.

In Australia, field trials have been prevented by an ongoing drought. A summer crop of sorghum was grown in 2005/2006 at Gatton (N responsive site) and Dalby (N and P responsive site) to assess N and P release from feed lot manure (FLM), green waste compost (GWC), and mixtures of these two wastes. Unfortunately, the soil profile contained considerable mineral N (150 kg/ha at Gatton, and 100 kg/ha at Dalby) as a result of mineralization during the long, drought-enforced, fallow period. The only significant difference observed was between 75 kg/ha fertilizer N application, and 20 t/ha GWC, the latter having a lower yield, presumably as a result of N immobilization by the GWC.

In addition to the field trials conducted to assess nutrient release from FYM, laboratory studies were undertaken to determine N release from organic materials with a wide range of C:N ratios (chickpea, soybean and wheat straw, Gliricidia, subabul, poultry manure and FYM). Degradation of low C:N ratio materials resulted in an immediate release of N, while the lower C:N materials produced an initial period of immobilization followed by mineralization. The addition of wheat straw (C:N ratio 79) produced a net immobilization throughout the 16 week incubation period. For FYM, the net immobilization period lasted for 4 to 6 weeks. At the end of the 16 week incubation period (equivalent to a crop growth season) only 31% of the FYM N had been mineralized. Total N content, and C:N ratio were the most effective predictors of N release. In Australia, N release from feedlot manure (FLM) and from green waste compost was assessed. As expected, N was readily mineralized from FLM, but even after 32 weeks incubation the GWC produced a net immobilization.

Integrated nutrient management (the combination of nutrient supply from organic matter and inorganic fertilizers) trials were conducted for a second year in Raigarh and Bhopal districts (two replicated trials in each district). Treatment regimes were developed to evaluate inorganic fertilization, organic fertilization, and a combination of inorganic fertilizer and organic fertilization for the soybean/wheat cropping system. In the soybean season, the treatments used included inorganic N, P, K, S and Zn at recommended rates, 5t FYM + 50% of the recommended rate of inorganic fertilizer, 5t farm yard manure (FYM) + 50% inorganic fertilizer + rhizobium inoculation, and an organic matter treatment of 5t FYM. Other treatments evaluated the effectiveness of phosphorus solubilising bacteria, and Azotobacteria (a free living N fixer). A farmers' practice treatment was also included.

The Kharif (monsoon season) soybean crop was adversely affected due to intense rainfall and flooding over much of Madhya Pradesh, and one of the projects experimental sites was lost. In the remaining three trials, the highest soybean grain yields were obtained in the 50% inorganic fertilizer + 5t FYM + rhizobium (2.0 t/ha), with the 100% inorganic fertilizer treatment (1.8 t/ha) as the second highest performing treatment. Similar results were obtained in 2005-2006, with the 100% inorganic fertilizer (2.1 t/ha), 50% inorganic + 5t FYM (2.2 t/ha), and 50% inorganic + 5t FYM + rhizobium inoculation (2.3 t/ha) producing yields which were not significantly different to each other, but significantly more than the other treatments tested (1.8 to 1.9 t/ha). An economic analysis (mean of two years 2005-06 and 2006-07) of the INM options in soybean revealed that the 50% inorganic fertilizer + 5 t FYM/ha + rhizobium to soybean produced the highest net returns (Rs. 13457 /ha) and highest benefit:cost ratio (1.28) to the farmer.

In the wheat season, these same treatments also produced the best grain yields of 4.9 to 5.1 t/ha (the 100% inorganic fertilizer treatment again receiving the recommended rates of N, P, K, S and Zn, and the integrated nutrient management treatments receiving 75% of the recommended rate of inorganic fertilizer). This result was also obtained in 2005-2006, with yields of 4.8 to 5.0 t/ha obtained. In both years, the organic fertilization treatments, which received 8t FYM prior to planting the wheat, and the farmers practice, produced significantly lower yields (around 4t/ha). Economic analysis of INM treatments in wheat showed that the100% inorganic fertilizer treatment produced the highest net returns (Rs. 48383 /ha) to the farmer, while the INM treatment, which only received 75% of the inorganic fertilizer produced the highest benefit:cost ratio of 4.4:1. As with soybean, both of these treatments produced higher net returns than the farmers' practice.

The INM option of 50% inorganic fertilizer + 5t FYM/ha + rhizobium to soybean and 75% inorganic fertilizer to wheat was found to be economically viable, and culturally attractive, nutrient management option for small farmers. These farmers can easily produce 5 tonnes of farmyard manure per annum, and this is adequate for their small land area. In case of medium (2-10 ha) and large (>10 ha) land holdings, it may not be possible to cover entire holding with the INM nutrient management strategy every year as the FYM available is not sufficient. These farmers are more likely to use the 100% inorganic fertilization approach on part of their holding to sustain higher productivity of soybean-wheat system. However, during group meetings farmers stress the value of FYM, and all farmers will continue its use, at least on part of their land. Thus, there is a high level of willingness to adopt the more effect INM approaches developed in this project.

In the initial year of the project, omission trials were conducted in farmers fields to determine if nutrients, other than nitrogen and phosphorus which are routinely used by farmers, are limiting yields. This activity demonstrated responses to zinc and sulfur. The nutrient omission trials proved to be an effective means of demonstrating these deficiencies, and farmers in the study area now realize the importance of these nutrients, and are applying gypsum or single superphosphate to supply sulfur, and zinc sulfate to supply zinc. The experimental approach of using nutrient omission trials at a farmer's field scale is relatively simple, and as the results are unambiguous, this is an effective extension tool. BAIF are interested in developing this approach further for more widespread use.

One of the primary aims of the project has been to overcome nutritional limitations in the soybean / wheat cropping system used in the monsoonal environment of Madhya Pradesh. To achieve this, the project team have used a series of experimental approaches. In the first year of the project, omission trials (where one plant nutrient at a time is omitted from the fertilizer mix) were used to identify those nutrients limiting the productivity of crops. In the second and third years of the project, fertilization regimes to address the nutrient deficiencies identified were developed and evaluated. Throughout the project, our aim has been to develop agronomic practices which are acceptable to farmers. To ensure this, there has been a high level of engagement with farmers; all experiments have been conducted on farmer's fields, and farmer's field days have regularly been conducted in order to understand farmer's perceptions of the work.

The project has developed two fertilizer management strategies. An Integrated Nutrient Management (INM) approach, where the use of farm yard manures (FYM) is combined with inorganic fertilizer. The traditional practice has been to apply a large application of manure (20 t/ha) infrequently. The research conducted in this project has demonstrated that substantial benefits can be gained from a smaller application (5 t/ha), permitting farmers to treat a larger area with manure each year. For the soybean crop the INM treatment consists of 50% inorganic fertilizer + 5t FYM + rhizobium, while the wheat crop receives 75% of the recommended rate of inorganic fertilizer. Even at this reduced rate of FYM application (5 t/ha), there is insufficient FYM for application to all of the cropped area, hence there is the need for an inorganic fertilizer regime termed Balance Fertilization (BF).

In the 2007 - 2008 year, the INM and BF were evaluated using a "Baby Trial" strategy. A total of 95 trials were established, distributed across 10 villages in three districts. These baby trials were conducted by farmers, with each trial consisting of three treatments; INM, BF, and farmer's practice. The aim was to confirm the experimental results obtained in researcher conducted trials, to assess how robust the technology was (would different farmers obtain comparable results), and to begin the process of extending these new fertilizer management practices to farmers. The 95 trial sites used were typical of farms throughout Madhya Pradesh, on the basis of soil testing, all were deficient in N, and for each of the nutrients P, S and Zn about 50% of sites were deficient - 32% of sites were deficient in all four of these nutrients.

In the Karif (monsoon season) soybean crop, inorganic fertilization produced yields 23% greater than the farmer's practice while the INM approach produced yields 46% higher than farmer's practice. During a farmers' field day, farmers attributed the higher soybean yield under INM to the better pod bearing relative to BF. The greatest differences between treatments were observed where that farmers maintained effective control of weeds, pests, etc, illustrating the impact of farm management on the effectiveness of fertilizer practice.

In the rabi (winter) season, wheat (variety Lok-1) was grown in the baby trial plots. Four crops out of the 95 planted could not be harvested, due to lack of irrigation water. Out of 91 harvested trials, 45 farmers used 3-4 irrigations, 12 farmers irrigated twice, and remaining 34 farmers could only irrigate once. The mean wheat grain yield of 91 trials showed that the INM produced a grain yield 24% higher than the farmers' practice, while BF increased the wheat grain yield by 30% over the farmers' practice. Yield differences were greatest where 3 to 4 irrigations had been used.

Much of the land suitable for cultivation in Madhya Pradesh is left uncultivated during kharif season because of waterlogging (e.g. 20-25% in Vidisha district). Even though some waterlogged fields are sown with soybean, the yields are very low due to poor establishment. The use of broad bed and furrow is being evaluated as an agronomic strategy to overcome some of the adverse effects of waterlogging. Trials during the last year showed that the integration of BBF with inorganic fertilization produced 40 to 50% higher soybean yield compared to a flat field with the same fertlizer regime. While this is a good result, it is important to note that the area only received about 70% of normal monsoon rainfall. Thus the yields on flat fields would be lower in a normal year, and the effectiveness of the BBF approach must be established in a wetter year. Nevertheless, the BBF approach has attracted the attention of the many farmers. Broad Beds formed during the kharif became almost flattened by the time of harvest of soybean. The wheat crop was grown on normal flat land in rabi season.

One of the primary aims of the project has been to overcome nutritional limitations in the soybean / wheat cropping system used in the monsoonal environment of Madhya Pradesh. To achieve this, the project team have used a series of experimental approaches. In the first year of the project, omission trials (where one plant nutrient at a time is omitted from the fertilizer mix) were used to identify those nutrients limiting the productivity of crops. In the second and third years of the project, fertilization regimes to address the nutrient deficiencies identified were developed and evaluated. Throughout the project, our aim has been to develop agronomic practices which are acceptable to farmers. To ensure this, there has been a high level of engagement with farmers; all experiments have been conducted on farmer's fields, and farmer's field days have regularly been conducted in order to understand farmer's perceptions of the work.
The project has developed two fertilizer management strategies. An Integrated Nutrient Management (INM) approach, where the use of farm yard manures (FYM) is combined with inorganic fertilizer. The traditional practice has been to apply a large application of manure (20 t/ha) infrequently. The research conducted in this project has demonstrated that substantial benefits can be gained from a smaller application (5 t/ha), permitting farmers to treat a larger area with manure each year. For the soybean crop the INM treatment consists of 50% inorganic fertilizer + 5t FYM + rhizobium, while the wheat crop receives 75% of the recommended rate of inorganic fertilizer. Even at this reduced rate of FYM application (5 t/ha), there is insufficient FYM for application to all of the cropped area, hence there is the need for an inorganic fertilizer regime termed Balance Fertilization (BF).
In the 2007 - 2008 year, the INM and BF were evaluated using a "Baby Trial" strategy. A total of 95 trials were established, distributed across 10 villages in three districts. These baby trials were conducted by farmers, with each trial consisting of three treatments; INM, BF, and farmer's practice. During 2008 - 2009 a reduced version of this experiment is being repeated by BAIF (using funds primarily provided by the farmers, and limited support provided by the ACIAR project of funds transferred from the Australian part of the project). This activity aimed to engage farmers not in the original group. During the final review of the project, the farmers from the original baby trial program indicated that they had adopted the improved practices - the particular nutrient management package adopted, depended on their circumstances (primarily, the availability of farm yard manure)
Much of the land suitable for cultivation in Madhya Pradesh is left uncultivated during kharif season because of waterlogging (e.g. 20-25% in Vidisha district). Even though some waterlogged fields are sown with soybean, the yields are very low due to poor establishment. The use of broad bed and furrow was evaluated in the project as an agronomic strategy to overcome some of the adverse effects of waterlogging. Trials during the last year of the project showed that the integration of BBF with inorganic fertilization produced 40 to 50% higher soybean yield compared to a flat field with the same fertlizer regime. While this is a good result, it is important to note that the area only received about 70% of normal monsoon rainfall. Thus the yields on flat fields would be lower in a normal year, and the effectiveness of the BBF approach must be established in a wetter year - this will be further evaluated in the extension year (2009-2010). Nevertheless, the BBF approach has attracted the attention of the many farmers. Broad Beds formed during the kharif became almost flattened by the time of harvest of soybean. The wheat crop was grown on normal flat land in rabi season.