Free, open access repository of research data produced as part of the Cereal Systems Initiative for South Asia (CSISA) project.
Cereal Systems Initiative for South Asia
Studies: 14

An on-farm experiment was carried out to evaluate mungbean machine sowing compared to conventional farmers’ tillage and crop establishment practices in Kharif-1 season with sowing from 28 January to 2 February 2015.

Eleven mungbean farmers, each of whom had at least five years of experience growing mungbean, were selected randomly at roughly 3 km intervals along a north to south transect in Patuakhali. This resulted in an approximate 33 km long stretch that was utilized in order to capture variation in microclimate that may alter yield, and to assess if power-tiller operated seeder (PTOS) performance is consistent across a landscape and variety of environments. In each farmers’ field, we imposed two treatments randomly in plots including conventional farmers’ practice (FP), which entailed tillage with a power tiller followed by broadcasting of seed and then incorporation of mungbean. The number of power tiller passes were decided upon by participating farmers. This control treatment was contrasted with single-pass PTOS seeding and fertilizing, with a target seed and fertilizer placement depth of 5 cm. BARI mungbean-6 was selected as an advanced variety for the experiment.

We collected possible agronomic data and constructed partial budgets for each treatment for farmer. Labor, tillage, fertilizer, seed, costs, resulting grain prices were determined by asking the farmers and the service providers. All data information aligned with metadata for further study.

Last Released: Nov 8, 2017

Under CSISA Phase II, Nutrient Omission Plot Technique (NOPT) trials were conducted in eight districts of Bihar, ten districts of Uttar Pradesh and ten districts of Odisha. Partner institutions include Bihar Agriculture University (BAU), Banaras Hindu University (BHU), Orissa University of Agriculture and Technology (OUAT), Central Rice Research Institute (CRRI), Odisha.

Last Released: Nov 8, 2017

The purpose of the study was

(1) to assess the economic viability of custom-hire service for ZT in CSISA's target districts in Bihar, the development of ZT service provision businesses over time, and constraints to further business expansion;

(2) to assess service providers’ resource endowment, risk exposure, risk preferences, and risk management practices;

(3) based on (2), to identify influencing factors of engaging in ZT service provision, including influencing factors of the scale of the service business operation.

38 downloads + analyses
Last Released: Nov 8, 2017

The purpose of the study was

(1) to assess the performance of ZT wheat as compared to conventional-tillage wheat in farmers' fields in six CSISA target districts in Bihar;

(2) to assess farmers’ resource endowment, risk exposure, risk preferences, and risk management practices;

(3) based on (2), to identify influencing factors of farmers' awareness and adoption of ZT in wheat, including social network effects.

Last Released: Nov 8, 2017
A Biophysical and Socioeconomic Characterization of the Cereal Production Systems of Northwest Bangladeshby Prabhakaran T. Raghu; Sreejith Aravindakshan; Frederick Rossi; Vijesh Krishna; Elahi Baksh; Azahar Ali Miah

This data was collected as part of the Cereal Systems Initiative for South Asia (CSISA) (Phase-I) project, with an objective to assesses cereal production in selected areas of NW-Bangladesh, especially with respect to the biophysical and socio-economic characterization of the cereal producing farm households.

Last Released: Nov 8, 2017
Nepal Rice Crop Cut & Survey Data 2016by Gokul Paudel; Shashish Maharjan; David Guerena; Ashok Rai; Andrew James McDonald

The major objectives of this rice crop cuts and survey were to:

(i) identify major rice yield determinants in western terai districts of Nepal and

(ii) to identify major determinants that describe the variability in rice yield across farms thereby linking with farm characteristics and socio-economic gradients.

This study uses Remote Sensing (RS) data particularly the NDVI (Normalized Difference Vegetative Index) value extracted from LandSat satellite images in each of the six districts (Kanchanpur, Kailali, Bardyia, Banke, Kapilbastu and Rupendhai). Since the resolution of LandSat satellite was better than MODIS; we used LandSat derived NDVI values to capture the variability in standing green biomass (a proxy for yield) so that sample selected will represent this variability in NDVI values. In each of the districts we saw a normal distribution of NDVI and samples were selected randomly by stratifying the NDVI values into four quartiles of bell curve, so that selected sample will represent proportionally the bell curve quartiles.

The study contains several optional modules questions and mandatory modules. The mandatory modules give the broader insights on input uses across farms while the optional module provides better insight on each input uses for example; fertilizer use details, irrigation dynamics, weed management and socio-economics gradients particularly based on income and expenditure of households. The optional module is accompanied by the crop cuts data which was done in three quadrants of each plot with 2*2=4 meter square area.

The sample size for this survey is 1052 households and the inputs use were asked for largest rice grown plots as farms may have multiple plots and inputs use might be different in different plots. Out of the total samples of 1052; a set of (~12%; 126 samples) households participated in optional modules to get the detail data on irrigation dynamics, weed management and socio-economics parameters.

Last Released: Nov 7, 2017
Maize crop cut data from Odisha plateauby Wasim Iftikar; Nabakishore Parida; Anurag Ajay

Maize crop cut data from farmer's field collected at Odisha plateau ecology.

Last Released: Nov 7, 2017
Untangling crop management and environmental influences on wheat yield variability in Bangladesh: An application of non-parametric approachesby Timothy J. Krupnik; Zia Uddin Ahmed; Jagadish Timsina; Samina Yasmin; Farhad Hossain; Abdullah Al Mamun; Aminul Islam Mridha; Andrew J. McDonald

In South Asia, wheat is typically grown in favorable environments, although policies promoting intensification in Bangladesh's stress-prone coastal zone have resulted in expanded cultivation in this non-traditional area.

Relatively little is known about howto best manage wheat in these unique environments. Research is thus needed to identify ‘best-bet’ entry points to optimize productivity, but classical parametric analyses offer limited applicability to elucidate the relative importance of the multiple factors and interactions that influence yield under such conditions. This problem is most evident in datasets derived from farmer-participatory research, where missing values and skewed data are common.

This paper examines the predictive power of three nonparametric approaches, including linear mixed effects models (LMMs), and two binary recursive partitioning methods: classification and regression trees (CARTs)and Random Forests

We collected yield, crop management, and environmental observations from 422 wheat fields in the 2012–13 season, across six production environments spanning southern Bangladesh, where nutrient rates and genotypes were imposed, but management of other production factors varied from farmer to farmer. Fields were grouped into categories including early and late-sowing, depending on crop establishment before or after December 15, respectively, and in combination, across both early- and late-sowing groups.

For each of these groups, we investigated how each non-parametric analysis predicted the factors influencing yield. All three approaches identified nitrogen rate and environment as the most important factors, regardless of sowing category. CART also identified assemblages of high- and low-yielding environments, although those located in saline and warmer thermal zones were not necessarily the lowest yielding, indicating that farmers can optimize crop management to overcome these constraints.

The number of days farmers sowed wheat before or after December 15, days to maturity, and the number of irrigations and weedings also influenced yield, though each method weighted these factors differently.

LMMs also indicated a slight yield advantage when farmers used stress-tolerant genotypes, though CART and Random Forests did not. One-to-one plots for observed vs. predicted yields from LMMs and Random Forests showed better performance by the former than the latter, with smaller root mean square and mean absolute error for the combined, early- and late-sowing groups, respectively.

While the LMMs were superior in this case, Random Forests may still prove useful in the classification and interpretation of farm survey data in which no treatment interventions have been administered.

Last Released: Nov 7, 2017
Odisha Rice Crop Cut Data 2013 - 2016by Wasim Iftikar; Nabakishore Parida; Vivek Kumar; Narayan Chandra Banik; Amit Mishra

Rice crop cuts conducted in Odisha (Mayurbhanj, Balasore, Keonjhar, Bhadrak, Khorda, Puri districts)

Last Released: Oct 23, 2017
Forgoing the fallow in Bangladesh’s stress-prone coastal deltaic environments: Effect of sowing date, nitrogen, and genotype on wheat yield in farmers’ Timothy J. Krupnik; Zia Uddin Ahmed; Jagadish Timsina; Md. Shahjahan; A.S.M. Alanuzzaman Kurishi; Azahar A. Miah; B.M. Saidur Rahman; Mahesh K. Gathala; Andrew J. McDonald

Working with 64 farmers ineight production environments, we examined yield response to three genotypes, BG25 and BG27 (with salinity - and heat - tolerant traits) and BG21 (local check), across a gradient of sowing dates, grouped as ‘early’ (sown before 15 December) and ‘late’ (after 15 December), under 0, 100 and 133 and 0, 67 and 100 kg N ha-1 for early- and late-sowing groups, respectively.

Across environments and genotypes, yield ranged from 2.11 to 4.77 t ha-1(mean: 3.9 t ha-1) under early-sowing, and from 0.83 to 4.27 t ha-1(mean: 2.74 t ha-1) under late-sowing. Wheat performance varied with environment (1.68 - 4.77 t ha-1 at 100 kg N ha-1across sowing groups); the lowest yields found where early sowing was delayed and soil salinity levels were elevated. Small but significant (P less than 0.001) yield differences (0.22 t ha-1) were found between 100 and 133 kg N ha-1 for the early-sowing group, though no difference was found between 67and 100 kg N ha-1 for late-sowing. Combining early- and late-sowing groups, significant environment x N rate and sowing-group x N rate interactions (both P less than 0.001) for 100 kg N ha-1 indicated the importance of site-and time-specific N management in these stress-prone environments.

Considering all cultivars and environments, ECa at sowing, flowering and grain filling negatively correlated with yield (r = - 0.50, - 0.59 and - 0.54, all P less than 0.001). Correlations with ground water depth at flowering and grain filling were negative and significant, but less pronounced in the context of farmer-managed irrigation scheduling. Despite putative stress-tolerance traits in two of the three entries, no genotypic yield differences were found under early-sowing, though small differences (less than 0.19 t ha-1) were observed with late sowing. Agronomic fertilizer-N efficiency (AE-N) was consistently higher for 100 than 133 and 67 than 100 kg N ha-1 for early- and late-sowing. The marginal economic value of N application followed similar trends, indicating that rates of at most 100 and 67 kg N ha-1 are favorable for sowing before or after December 15th.

Last Released: Sep 27, 2017