Levelling the farm fields: A cross-country study of the determinants of gender-based yield gaps

CONTEXTRecent studies on yield gap analysis for rice in Southeast Asia revealed different levels of intensification across the main ‘rice bowls’ in the region. Identifying the key crop management and biophysical drivers of rice yield gaps across different ‘rice bowls’ provides opportunities for comparative analyses, which are crucial to better understand the scope to narrow yield gaps and increase resource-use efficiencies across the region. OBJECTIVEThe objective of this study was to decompose rice yield gaps into their efficiency, resource, and technology components and to map the scope to sustainably increase rice production across four lowland irrigated rice areas in Southeast Asia through improved crop management. METHODSA novel framework for yield gap decomposition accounting for the main genotype, management, and environmental factors explaining crop yield in intensive rice irrigated systems was developed. A combination of crop simulation modelling at field-level and stochastic frontier analysis was applied to household survey data to identify the drivers of yield variability and to disentangle efficiency, resource, and technology yield gaps, including decomposing the latter into its sowing date and genotype components. RESULTS AND CONCLUSIONThe yield gap was greatest in Bago, Myanmar (75% of Yp), intermediate in Yogyakarta, Indonesia (57% of Yp) and in Nakhon Sawan, Thailand (47% of Yp), and lowest in Can Tho, Vietnam (44% of Yp). The yield gap in Myanmar was largely attributed to the resource yield gap, reflecting a large scope to sustainably intensify rice production through increases in fertilizer use and proper weed control (i.e., more output with more inputs). In Vietnam, the yield gap was mostly attributed to the technology yield gap and to resource and efficiency yield gaps in the dry season and wet season, respectively. Yet, sustainability aspects associated with inefficient use of fertilizer and low profitability from high input levels should also be considered alongside precision agriculture technologies for site-specific management (i.e., more output with the same or less inputs). The same is true in Thailand, where the yield gap was equally explained by the technology, resource, and efficiency yield gaps. The yield gap in Indonesia was mostly attributed to efficiency and technology yield gaps and yield response curves to N based on farmer field data in this site suggest it is possible to reduce its use while increasing rice yield (i.e., more output with less inputs). SIGNIFICANCEThis study provides a novel approach to decomposing rice yield gaps in Southeast Asia's main rice producing areas. By breaking down the yield gap into different components, context-specific opportunities to narrow yield gaps were identified to target sustainable intensification of rice production in the region.

Yield gap is an important aspect as it affects production. There is a need to take up in depth analysis of yield gap for narrowing down the yield gap between the farmers fields and demonstration plots. It may not always possible for the farmers to raise the crop productivity on their farm to the level of research station. However, it would be realizable to aim at demonstration plot yield. Hence in this study more emphasis is given to yield gap II. i.e. difference between demonstration plot yield and farmers field. The study was based on the secondery data collected from Agricultural Prices Scheme, Dr. PDKV, Akola for the year 2014 -15. In all 120 Arhar growing farmers from Akola, Buldhana and Amaravati district were selected. It is observed that total Yield gap was highest in low adopter (6.81 qt/ha) and less in high adopter (1.87 qt)/ha.

Yield gap (Yg) analyses using farmer-reported yield and management data have been performed for a number of annual grain crops, but it lacks for perennial forages. The U.S. accounts for 21 % of the global alfalfa production with a large rainfed area located in the central Great Plains, serving as an interesting case-study for Yg in perennial forages. Most existing alfalfa Yg analyses quantified the magnitude of the Yg but failed to identify associated management practices to reduce it. Challenging this analysis, a systematic benchmark for alfalfa water productivity [WP, kg dry matter per mm evapotranspiration (ETc)] that allows for the quantification of Yg in farmer fields does not exist. Our objectives were to (i) benchmark alfalfa WP, (ii) quantify Yg in alfalfa farmer fields, and (iii) identify management opportunities to improve alfalfa yield. We conducted a systematic review of literature and compiled a database on alfalfa yield and ETc (n = 68 papers and 1027 treatment means) from which a WP boundary function was derived. We collected management and yield data from 394 commercial rainfed alfalfa fields during 2016–2019 in central Kansas. We then used satellite imagery to define the growing season (and corresponding water supply) for each field. The boundary function was then used to calculate Yg of each field, and conditional inference trees (CIT) explored the impact of management practices associated with increased yield. Our boundary function suggested an alfalfa WP of 34 kg ha-1 mm-1. Farmer-reported yield ranged from 0.9 to 19.0 Mg ha-1, averaging 7.6 Mg ha-1. Alfalfa water-limited yield potential (Yw) ranged from 11.1 to 23.2 Mg ha-1, resulting in an average yield gap of 54–60 % of Yw. Row spacing, seeding rates, and management of phosphorus fertilizer were major agronomic practices explaining alfalfa yields in farmer fields, followed by surrogate variables as sowing season, stand age, and soil pH. Our study provided the first systematic analysis estimating attainable alfalfa WP as function of ETc, suggesting that large alfalfa Yg exist in the U.S. central Great Plains. We also identified key agronomic practices associated with increased alfalfa yield. The WP here derived can be used for future studies aiming at quantifying alfalfa Yg across the globe. This was an initial step in quantifying Yg and its associated causes at farmer fields, and we highlight limitations and future directions for perennial forages yield gap analyses.

Yield gap analyses are useful for examining differences between potential yield (Yp) and actual yield realized in farmer fields. However, it is often difficult to estimate the attainable yields in smallholder farm agriculture. In this study, we estimated the yield gaps in irrigated wheat in the Hebei Province in China using data from farmer surveys and crop modeling. The EPIC (Environment Policy Integrated Climate) model was calibrated and evaluated for high‐yielding wheat experiments in Hebei Province to estimate the Yp at the county level. The model was run for 10 yr (2005–2014) using county‐level weather data, and the area‐weighted average model‐based estimate of Yp was 9.93 Mg ha−1 across 116 counties in Hebei Province, which was higher than the estimate of approximately 9.0 Mg ha−1 that was reported in previous studies. The attainable yield potential (Ya) during this period was defined as the mean yield of the top 5% of farmers, and the actual farm yield (Yf) in these counties was calculated based on a survey of 385,167 farmers. The area‐weighted average Yf was 6.47 Mg ha−1, corresponding to 65.2% of the model‐based estimate of Yp. The total yield gap was high at 3.46 Mg ha−1. The top‐producing farmers achieved only 77% of the model‐based Yp, which indicates that closing the yield gap remains a challenge. In the future, investments should be made in technology transfer for irrigated wheat production in China.Core Ideas We evaluated yield gap based on model study and survey of 385,167 farmers. Model‐based yield potential of irrigated wheat averaged 9.93 Mg ha−1. Farmers achieved 65.2% of yield potential, giving a yield gap of 3.46 Mg ha−1. The top 5% of farmers in the survey achieved 77% of yield potential.

The main purpose of the research was to identify factors responsible for yield gap in wheat production. Eighteen (18) experiments were conducted in two major wheat growing districts Rangpur and Dinajpur in two consecutive years. The selected varieties for the conducted research were Prodip, Satabdi and Sourav. All the experiments were established in farmers’ fields providing all recommendations for wheat production. It was observed that yield gap varied with the variety and farmers to farmers and location to location. The overall yield gap of Prodip was the highest (18.43 percent) followed by Sourav (18.15 percent) and Satabdi (17.45 percent). Yield gaps of all the wheat varieties under study were higher in Rangpur site than Dinajpur site. The practice gap was the highest in gypsum application (69 percent) followed by boron (67 percent), sowing time (40 percent). Practice gaps in the application of MoP, TSP and irrigation were almost equal, 40 percent, 37 percent, and 36 percent respectively. Late sowing, non use of dolomite and micro nutrients (zinc and boron) in wheat yield with sub-optimal doses of phosphatic and potash fertilizers were the main reasons for yield gap. Adoption of short duration T. aman variety and optimal doses of chemical fertilizers with micro nutrients in wheat field could minimize this gap to a greater extent. Preventive measures against bird attack after sowing of seeds for optimum plant population would have impact in narrow down this yield gap as well.DOI: http://dx.doi.org/10.3329/pa.v23i1-2.16620Progress. Agric. 23(1 & 2): 91 – 99, 2012

When yield gaps are poverty traps: The paradigm of ecological intensification in African smallholder agriculture

The present study was conducted at farmers’ field in Mathura and Ghaziabad districts of Uttar Pradesh and Bahadurgarh and Gurgaon districts in Haryana during 2010 to 2012. The results of micro yield gap analysis from a sample size of 120 farmers revealed that the average yield gap-I (technology gap) for pigeonpea was 1167 kg/ha in Uttar Pradesh and 1250 kg/ha in Haryana. While, the average yield gap-II (extension gap) for pigeonpea was relatively lower i.e. 183 kg/ha in Uttar Pradesh and 125 kg/ha in Haryana. The average yield gap-I for chickpea was observed as 1641 kg/ha in Uttar Pradesh and 877 kg/ha in Haryana. Whereas, the average yield gap-II for chickpea was relatively lower i.e. 614 kg/ha in Uttar Pradesh and 622 kg/ha in Haryana. However, there was lower yield gap for paddy and wheat in both the states i.e. Uttar Pradesh and Haryana in comparison to pulses (pigeonpea and chickpea). Therefore, it is summarized that technology gap in pulses (pigeonpea and chickpea) was more than extension gap at farmers field. The potential interventions and various constraints of yield gap in major pulses have been highlighted in this paper.

An attempt has been made to study on impact of demonstration on farmers fields in adopted villages of Sikar district. The yield gap, input gap, cost and return were calculated for purpose of the study. The survey covered 40 farmers from 4 adopted villages where wheat, barley, gram and mustard demonstration (full technology) were conducted in adopted villages. In each village, 10 demonstrations were conducted. From each village, 10 farmers were selected who have adopted traditional practices for crop cultivation. The results indicated that yield gap per hectare between demonstration plots and farmers practices was 15.71, 23.70, 29.17 and 20.00 per cent for wheat, barley, gram and mustard, respectively. On farmers practices, overall inputs gap was about 21 per cent for wheat, barley and mustard and 33 per cent for gram as compared to demonstration. Thus, there is more scope to raise the mustard, gram, barley and wheat productivity by improving the techniques of production rather than by raising the input use levels. The results further revealed that the cost of cultivation per hectare on demonstration plots was Rs. 16854, Rs. 15110, Rs. 13622 and Rs. 12415 for wheat, barley, gram and mustard while on farmers fields it was Rs. 13883, Rs. 12445, Rs. 10301 and Rs. 10227 for wheat, barley, mustard and gram, respectively. The net return per hectare was the highest for mustard followed by wheat, barley and gram. While on farmers practices, it was highest for wheat followed by mustard, barley and gram. The increase of net return on demonstration plots over farmers' practices was 26.79, 24.75, 19.25 and 9.43 per cent for barley, gram, mustard and wheat, respectively.

Understanding the reasons for the yield gap between potential and actual yield can provide insights for enhancing canola production by adapting measures for ensuring food security. The canola yield gap under different management practices (e.g. water, nitrogen, N- and sowing dates) was quantified using research trials that were conducted at on-station and historical data (1980–2016) and the CROPGRO-Canola model for Punjab, Pakistan. The integrated approach revealed that low inputs of N, the amount of irrigation, sowing date and the use of seeds from home stocks were the principal causes for a low yield. The CROPGRO-Canola model was able to simulate the canola yield from research trials (R2 = >0.90) and farm survey data (R2 = 0.63). The average yield gap between potential (YP), N-limited (YNL), water-limited (YWL), N- and water-limited (YNWL), and overall farmer field yield (YOFF) was 50, 46, 62 and 72%, respectively. The yield-gap with achievable yield (YA) for YNL, YWL, YNWL and YOFF was 34, 28, 49 and 63%, respectively. Overall, the results showed that a high canola yield for farmers’ fields can be obtained by selecting appropriate varieties and sowing dates with N rate of 120 kg/ha and efficient irrigation management. However, further studies are necessary to fully comprehend the underlying causes for the low actual yield and the high yield variability of farmers’ fields.

The present study was conducted at farmers’ field in Narsinghpur and Umaria districts of Madhya Pradesh; Wardha and Yavatmal districts in Maharashtra during 2016 to 2017. The results of yield gap analysis from a sample size of 160 farmers revealed that the average yield gap-I (technology gap) for pigeon pea and chickpea was 712 to 817 kg/ha and 755 to 789 kg/ha in Madhya Pradesh and in Maharashtra 500 to 657 kg/ha and 395 to 627kg/ha. While, the average yield gap-II (extension gap) for pigeon pea was relatively lower i.e. 426 to 448 kg/ha in Madhya Pradesh and 454 to 558 kg/ha in Maharashtra. Whereas, the average yield gap-II for chickpea was relatively lower i.e. 264 to 421 kg/ha in Madhya Pradesh and 427 to 518 kg/ha in Maharashtra. However, the overall yield gap analysis in pulses in both the district of Madhya Pradesh found that technology gaps (gap-I) were observed more than extension gap (gap-II) in varieties of both the crop. In case of both the district of Maharashtra found that technology gaps (gap-I) were observed less than extension gap (gap-II) in varieties of both the crops except variety Jaki 9218 of chickpea and ICPL 8863 variety of pigeon pea. Therefore, it is summarized that technology gap in pulses (pigeon pea and chickpea) was more than extension gap at farmers field. The potential interventions and various constraints of yield gap in major pulses have been highlighted in this paper.

Quantifying the yield gap in wheat–maize cropping systems of the Hebei Plain, China

Rice yield gaps and nitrogen-use efficiency in the Northwestern Indo-Gangetic Plains of India: Evidence based insights from heterogeneous farmers’ practices

Yield, yield gaps, input use, N‐use efficiency, productivity, and profitability of irrigated rice in Burkina Faso were determined for a typical irrigation scheme in the dry season (DS) 1999 and the wet season (WS) 2000. Objectives were to analyze agro‐economic constraints and opportunities and determine ways to overcome such constraints. The simulation model RIDEV was used to analyze farmers' crop management practices, revealing considerable deviation between actual and optimal timing of crop management interventions. This diversity of cropping practices caused considerable variation of internal efficiency of N (IEN), partial factor productivity of N (PFPN), N recovery fraction (RFN), rice (Oryza sativa L.) grain yields, and net benefits of N‐use. The results showed a clear relation between plant N uptake and yield (mean IEN of 48 kg grain kg−1 N uptake in farmer's fields), but the relation between N applied and yield was scattered. The PFPN varied from 16 to 52 kg grain kg−1 N applied (mean of 35 kg grain kg−1 N applied) due to a large range of fertilizer N recovery rates (RFN = 7–77%; mean of 37%). Farmers' average yields were 4.9 Mg ha−1 in the DS and 3.6 Mg ha−1 in the WS, but yields were very variable and ranged from 0.9 to 7.9 Mg ha−1 in the DS and from 1.0 to 7.9 Mg ha−1 in the WS. Yield gaps between average farmer's yield and best farmer's yield were high (3.0 Mg ha−1 in the DS and 4.3 Mg ha−1 in the WS), indicating considerable scope for yield increases in both seasons. Net benefits to irrigated rice cropping were mostly positive (avg. $418 (US) ha−1) in the dry season, but very low in the wet season (avg. $108 (US) ha−1). Partial budget analysis of fertilizer use revealed considerably lower value/cost ratios of fertilizer use in the wet season (mean V/C: 1.5) compared with the dry season (mean V/C: 2.9). It was concluded that improved crop management practices are the key to reach higher yields and financial returns without additional investments.

Krishi Vigyan Kendra Yachuli, Lower Subansiri district, gave a front-line field pea demonstration in 06 villages across two blocks with 44 farmers during the 2019-20 and 2020-21 seasons. FLD on the VL Matar 42 variety of field pea was carried out over the course of two years in a 15 ha area using the recommended improved practices. Additionally, a control plot with farmer practices was maintained. The yield in the farmers' plot (1110 kg ha-1) and demonstration plot (1510 kg ha-1) in the year 2020-21 was higher than in the year 2019-20. In the years 2019-20 and 2020-21, respectively, the demonstration plots' mean yield exceeded that of the farmers' plot by 35.23 and 36.03 percent. The VL Matar 42 variety of field pea had a mean yield of 1465 kg ha-1, which was lower than the potential yield of 1868 kg ha-1. The yield gap of 403 kg ha-1 indicates that there is a technology gap. Interestingly, the average extension yield gap was lower (385 kg ha-1) during the study period. The technology index varied from 19.16 to 23.98 percent, showing the feasibility of the evolved technology at the farmer's fields. Cultivating field pea using improved technologies resulted in an average higher net return of Rs. 59,050 ha-1 compared to Rs 34,500 ha-1 from local farming practices. The benefit cost ratio of field pea was higher (2.16) when using improved technologies compared to (1.83) when using farmers' practices.

Improved agronomic management is important to reduce yield gaps and enhance food security in sub-Saharan Africa. This study was undertaken to understand contributing factors to observed yield gaps for maize in farmer fields and to demonstrate appropriate agronomic survey methods. The study aimed to (1) demonstrate an approach for farm-level agronomic survey, (2) identify key crop production constraints and (3) define the nutrient input and output balances of different fields. Agronomic survey was conducted in 117 farmer fields randomly distributed in a 10 km by 10 km block in Babati, northern Tanzania. A semi-structured questionnaire and production measurements were used to collect data which were analyzed with regression classification and mixed effect models. The exploitable maize yield gap at farm-level reaches up to 7.4 t ha−1, and only <5 % of fields achieve maize grain yield of 5 t ha−1. Slope, plant density, distance from homestead, crop variety, timing of planting and period since conversion significantly influenced maize yields. For example, fields on flat land had up to 1.6 t ha−1 more maize grain yield than those on steep slopes while fields with plant density >24,000 plants ha−1 had 900 kg ha−1 more yield than those with less density. At least 52 % of the fields had negative nutrient balances. We conclude that cropping systems used in Babati should be preferentially supplemented with mineral fertilizers while optimizing plant density, increasing manure application and appropriate varietal choice in order to reduce the yield gaps.