Smallholder Fields Research Articles

Sources of Technical Inefficiency for Smallholder Field Pea Producers in Ethiopia

This study intended to measure the technical efficiency and sources of inefficiency for field pea producers in Ethiopia. The study used primary data collected from 207 smallholder farmers. A multi-stage sampling technique was followed to select sample households. A stochastic frontier was the model used, and a single-step estimation approach was followed. The result indicated that the mean technical efficiency is 49.23 %, and the average productivity of the crop is 0.96 tons/hectare. This indicates that the technical efficiency could be increased by 51.77 % through proper utilization of production variables, i.e. agrochemicals, oxen power, plot size, and seed rate. The inefficiency model result also confirmed that the technical inefficiency of field pea production was negatively affected by access to off-farm income, access to credit services, extension contacts, adoption of new technologies, access to training, and participation in contract farming of the crop. Therefore, this study recommends agricultural policies favoring vocational training to farmers on the new agricultural technology packages, availing access to off-farm incomes, facilitating credit services to farmers, and establishing legal frameworks that encourage contract farming.

Adopting yield-improving practices to meet maize demand in Sub-Saharan Africa without cropland expansion

Maize demand in Sub-Saharan Africa is expected to increase 2.3 times during the next 30 years driven by demographic and dietary changes. Over the past two decades, the area cropped with maize has expanded by 17 million hectares in the region, with limited yield increase. Following this trend could potentially result in further maize cropland expansion and the need for imports to satisfy domestic demand. Here, we use data collected from 14,773 smallholder fields in the region to identify agronomic practices that can improve farm yield gains. We find that agronomic practices related to cultivar selection, and nutrient, pest, and crop management can double on-farm yields and provide an additional 82 million tons of maize within current cropped area. Research and development investments should be oriented towards agricultural practices with proven capacity to raise maize yields in the region.

Accurate measurement of field size is essential for analysis of smallholder survey data

ContextAgronomic surveys play a vital role in identifying on-farm yield constraints and opportunities to increase productivity, input-use efficiency, and farmer profit. These surveys typically rely on farmer-reported data, where crop yield and applied inputs per hectare are estimated based on farmer-reported total harvested output and applied inputs per field and associated field size. At question is how accurate field size is, which is vital in the case of smallholders as small inaccuracies can lead to large biases in estimated yield and inputs. ObjectivesThe goals of this study are to assess current approaches to determine field size in smallholders, analyze biases between measurements and farmer estimates, and their influence on the analysis of the survey data. Cost-effective approaches for accurate field size measurement are also discussed. MethodsWe reviewed published studies that used survey data collected from smallholders to determine the extent to which field size is independently validated by researchers. Subsequently, we used data from thousands of smallholder fields in Southeast Asia to compare farmer-reported versus measured field size and examined the associated impact on estimated yields, inputs, and their relationships. ResultsWe found that most published studies using survey data did not validate field size reported by farmers. Furthermore, our analysis of the field size data from smallholders in Southeast Asia showed substantial differences between farmer-reported and measured field size, resulting in significant biases in estimated yield and applied inputs, and leading to spurious relationships between them. ConclusionUncertainty in field size can mislead analysis of smallholder survey data. ImplicationsFuture studies relying on survey data from smallholders should incorporate measurements of field size for accurate determination of crop yield and inputs.

Recovery status of coastal vegetation in Mozambique: Insights from floristic composition and soil organic carbon analysis

The growing demand for land for slash-and-burn cultivation, driven by human population growth, has led to the creation of scattered cultivated fields and a mosaic of fallow areas in various stages of regeneration. While tropical forests have demonstrated the ability to rebound regarding tree species diversity and forest structure following disturbances, the extent of recovery in coastal vegetation in southern Africa remains uncertain and poorly understood. To address this knowledge gap, our study collected biophysical data from smallholder fields, fallow land, and forests to evaluate the structure and functioning of coastal vegetation in Gaza province, southern Mozambique. We established 90 sample plots of 0.1 ha each at different stages of plant succession using opportunistic sampling approach. Within these plots, we recorded the diameter at breast height, total height, and species name of all trees and shrubs. Litter and soil samples were collected in 1 m² quadrats embedded within the main plot. Our findings indicate a high capacity for regeneration following the disturbance caused by slash-and-burn cultivation. Species composition varied significantly across different stages of succession, with 52 species identified in forests (FL), 34 in old fallow (OF), 29 in middle fallow (MF), 28 in initial fallow (IF), and 15 in cropland (CL). Richness and diversity increased over time after the practice of slash-and-burn cultivation, with a noticeable enhancement from the initial fallow stage. We observed a gradual recovery of soil carbon stocks in young fallow areas, while litter biomass decreased throughout the recovery phase of the fallow lands. Our study highlights that areas abandoned after agriculture can regenerate in terms of structure, floristic composition, and tree carbon content.

Detecting and quantifying zero tillage technology adoption in Indian smallholder systems using Sentinel-2 multi-spectral imagery

Zero tillage (ZT), an important component of Conservation Agriculture, has enormous potential to curb emissions from residue burning, increase soil organic carbon and water retention, reduce land preparation costs and increase the long-term productivity and profitability of the farming system. Despite the promise of ZT, little is known about how widely it has been adopted at regional scales in smallholder systems, where management is heterogeneous. Identifying ZT diffusion patterns across space and time along with other popular tillage technologies, such as conventional tillage (CT) and shallow tillage (ST), helps to target and disseminate the most effective technologies and estimate their climate change mitigation potential. Acknowledging the complexities involved in distinguishing ZT from CT and ST in smallholder fields, this study utilized an innovative two-step change detection method leveraging early-season Sentinel-2 multi-spectral imagery. We developed and applied our model in the Indian state of Punjab over three years (2020–2022). Our method outperformed traditional binary classification models, achieving 81 % accuracy. The analysis indicated that areas under different tillage types changed over time across Punjab. Specifically, from 2020 to 2021, we found a 33 % and 4 % decrease in ZT and CT, respectively. However, a 29 % increase is observed in CT adoption. On the other hand, from 2021 to 2022, the adoption rates for ZT and CT increased by 18 % and 2 %, respectively, while ST adoption decreased by 12 %. Overall, this study demonstrates the potential use of early season Sentinel-2 imagery to map the adoption of tillage practices in smallholder systems. Our approach can provide large-scale information on technology uptake, aiding policies to implement carbon markets and the scaling up of sustainable agricultural practices in India.

Fields of contestation and contamination: Maize seeds, agroecology and the (de)coloniality of agriculture in Malawi and South Africa

Farmer-managed seed systems and the conservation of agrobiodiversity are increasingly recognized as important components of food and seed sovereignty. In contrast, hybrid, genetically modified (GM), and, increasingly, gene-edited crops continue to be promoted by Green Revolution proponents as a “climate smart” package that includes fertilizers, pesticides, purchased seeds, and links to global markets. Influencing seed laws and policies to support the uptake of modern crop varieties has been a key entry point in many countries, facilitated by networks of foreign donors, philanthropists, governments, and multinational companies. Using the case of South Africa, where GM crops have been grown for several decades, we provide insights on implications for Malawi, which passed a Seed Act in 2022, implicitly supporting GM crops. Both countries have histories of colonial agriculture with strong policy support for modern, hybrid varieties of maize, and the replacement (and displacement) of local, open-pollinated maize varieties. In South Africa, several studies have revealed the contamination of smallholder fields and seed systems. Through a political ecology lens, we explore how maize and its cotechnologies were commodified in South Africa and Malawi, and what South African experiences of GM crop adoption in smallholder farming systems can tell us about the challenges to be faced by smallholder Malawian farmers. We reveal how colonial histories and ongoing colonialities of power, knowledge, being, and nature continue to shape the character and form of agriculture in both countries, running counter to the needs of agroecological smallholder farmers and their ways of knowing and being. We conclude by envisioning what reimagined, transformed and decolonial approaches for food and agriculture might look like on the African continent, and how they might contribute toward the attainment of food and seed sovereignty and an agroecological future.

Improved prediction of rice yield at field and county levels by synergistic use of SAR, optical and meteorological data

Timely and accurate rice yield prediction over large regions is imperative to making informed decisions on precision crop management and ensuring regional food security. Previous studies on crop yield prediction mainly used of either optical or synthetic aperture radar (SAR) data alone, and few of them investigated the synergistic use of these two data sources. It remains unclear whether and how synergistic use of optical and SAR imagery would improve crop yield prediction over a large region. In addition, various machine learning (ML) algorithms have been used in regional yield prediction, but the existing findings are mostly inconsistent about the optimal ML algorithm that prevails others across years and spatial levels. To solve these issues, this study designed a meta-learning ensemble regression (MLER) framework and developed an efficient method of integrating multi-source data for accurate prediction of rice yield at field and county levels over Jiangsu Province of China.The results demonstrated that SAR data from Sentinel-1 could compensate for the lack of optical data from Sentinel-2 for rice yield prediction, which was further improved by adding the meteorological data. The MLER algorithm exhibited the highest field-level prediction accuracy for Xinghua and Suining (R2 = 0.89, RMSE = 0.54 t/ha), as compared with individual base algorithms (RF: R2 = 0.80, RMSE = 0.72 t/ha; XGBoost: R2 = 0.76, RMSE = 0.80 t/ha; SVR: R2 = 0.74, RMSE = 0.83 t/ha) and the long short-term memory (LSTM) (R2 = 0.76, RMSE = 0.80 t/ha) with a ten-fold cross-validation. Furthermore, integrating the MLER algorithm with the best-performing data combination could reduce the influence of regional phenological variability for building yield prediction models applicable to the entire province. The MLER models for predicting the rice yield across the province exhibited robust performance in the leave-one-year-out accuracy assessment at both levels (field level: R2: 0.50∼0.61, RMSE: 0.85∼1.22 t/ha; county level: R2: 0.42∼0.67, RMSE: 0.23∼0.27 t/ha) and could be transferred to an extreme-heat year (2022) (R2 = 0.49, RMSE = 1.52 t/ha). This research opens up new possibilities for predicting the rice yield of smallholder fields over large regions at multiple levels by integrating publicly available multi-source data and machine learning.

Shortening harvest interval, reaping benefits? A study on harvest practices in oil palm smallholder farming systems in Indonesia

CONTEXTSmallholders are responsible for a large share of global palm oil production. Yet, in Indonesia, the main palm oil producing country, smallholders' yields remain low. Better management practices, including short harvest interval (HI, the number of days between two harvest rounds), could help to raise smallholder yields. However, at present, HI is long in smallholder fields and the drivers underlying this phenomenon are poorly understood. OBJECTIVEWe explored agronomic, socio-economic, and institutional factors that underlie harvesting practices in independent oil palm smallholder farming systems in Indonesia to assess scope for sustainable intensification through shorter HI and reduced harvest losses. METHODSCombining methods from agronomy and anthropology, we followed harvest interval of 950 farmers in six representative locations across Indonesia via farmer diaries over a period of two years to establish a correlation with yield. To quantify this relationship, we conducted post-harvest field measurements, and to explain which underlying factors impact HI we did qualitative interviews and surveys. RESULTS AND CONCLUSIONSThe HI of smallholders in our study ranged from 10 to 39 days (average: 17-d). Half of the farmers followed long HI (>16-d). Key factors impacting HI include annual fresh fruit bunch (FFB) yield, total palm area per farmer, trusted labor availability, plantation accessibility, and FFB price. Farmers responded to low yield by prolonging HI to increase labor productivity and optimize labor and transportation costs. SIGNIFICANCEThis study contributes to a better understanding of the relation between HI and yield in smallholder farming systems, by uncovering how socio-economic and institutional factors sometimes override agronomic considerations. Long HI can potentially lead to harvest loss from loose fruits and missed bunches, and reduce oil quality from overripe bunches. However, to obtain the benefits of shorter HI requires collective action and incentives along the supply chain to streamline the harvest and sale process.

Limited environmental and yield benefits of intercropping practices in smallholder fields: Evidence from multi-source data

To ensure food security in sub-Saharan Africa, it is necessary to improve crop yields while minimizing environmental impacts. Intercropping has been demonstrated to deliver such outcomes, but their performance in smallholder fields has received limited attention therefore insufficient to capture the complexity of real-world crop fields run by smallholder farmers. This study examines the benefits and management of intercropping practices in real smallholder fields in Malawi. We collected field data on intercrop types, the number of intercropped species and maize yield in intercropped maize fields. Field data was then combined with geospatial and household survey data to investigate the yield benefits, agricultural inputs, and factors related to intercropping choices. We used Pearson correlation and Tukey’s test to test the statistical significance in the difference between intercropped fields and monoculture fields. We found that more intercrops were planted in fields with smaller sizes, drier conditions, and higher soil erosion levels, with adoption rates increasing from 75 % in 2010 to 84 % in 2020. In addition, our field data shows that intercropping is associated with reduced primary maize yield (2.7 t/ha) compared to pure maize yield (3.8 t/ha). Conversely, satellite data demonstrates an improvement in overall field yield in intercropped fields. Meanwhile, intercropped fields require higher labor inputs (11 h more per season) and increased weeding times than monocultures, however agrochemical inputs (fertilizers and pesticides) do not necessarily decrease in intercropped fields compared to monocultures. Our results suggest that while smallholder farmers in Malawi adopt intercropping to improve land use efficiency, drought resilience, and soil fertility, they are not realizing the full benefits observed in experimental trials. More evidence on the benefits and best practices of intercropping in smallholder fields is necessary in order to better understand this practice as an option for sustainable intensification.

Too little, too imbalanced: Nutrient supply in smallholder oil palm fields in Indonesia

CONTEXTSmallholders account for ca. 40% of oil palm area in Indonesia but average yield remains low. Despite higher overall fertilizer use in Indonesia compared with other Southeast Asian countries, poor plant nutrition has been identified as a major factor explaining yield gaps in smallholder oil palm fields and little is known about the underlying management drivers. OBJECTIVETo assess nutrient management in smallholder fields and identify entry points to narrow the existing yield gap via improved plant nutrition. METHODSWe assessed nutrient balances and gaps for nitrogen (N), phosphorous (P), potassium (K), and magnesium (Mg) and inter-relationships between fertilizer use, leaf nutrient concentration, and yield, using data collected from 977 smallholder fields in Indonesia over two years. RESULTS AND CONCLUSIONSThere was a positive relationship between yield and nutrient rates, especially for potassium (K). Only half of the fields received fertilizer and average nutrient rates in fertilized fields represented 45% (N), 83% (P), 28% (K), and 25% (Mg) of the associated removal with harvested yield. Applied fertilizer was generally rich in N and P, but poor in K and Mg, leading to nutrient imbalances. Additionally, farmers did not follow appropriate practices on fertilizer placement and splits. Improving current fertilizer supply and management is needed to increase smallholder yields and profit. SIGNIFICANCECurrent agricultural and research programs will benefit from re-orienting efforts to improve nutrient management in oil palm as a pathway to narrow yield gaps in smallholder fields, which, in turn, requires strengthening extension services to fill in knowledge gaps and tune subsidy programs to facilitate farmer access to fertilizer sources more suited to the crop.

First things first: Widespread nutrient deficiencies limit yields in smallholder oil palm fields

CONTEXTIndonesia is the most important oil palm producing country. Nearly 40% of planted area is managed by smallholders, with yields well below the potential. Efforts to increase productivity have focused on the source of planting material, with little attention paid to plant nutrition. OBJECTIVETo assess the degree to which current productivity in smallholder oil palm fields is limited by nutrients in scenarios with distinct sources of planting material. METHODSWe collected detailed data on leaf nutrient concentration from 30 fields to derive minimum sampling size needed to diagnose nutrient status. Subsequently, we collected data on yield and palm type from 973 smallholder fields to assess the importance of nutrient status and planting material in the determination of yield. RESULTS AND CONCLUSIONSPotassium (K) deficiency was widespread (88% of fields) and often severe. Nearly two thirds of fields were also deficient for nitrogen (N) and boron (B), half were phosphorous (P) deficient, and one third were magnesium (Mg) deficient. Nutrient imbalances, especially between K and N, were also common. Fields with sufficient N, P, and K levels yielded 47% more (equivalent to 1.2 t oil ha−1) than deficient fields across the entire range of planting materials. We conclude that improved plant nutrition increases fresh fruit yields in smallholder fields irrespective of the source of planting material. The advantage of certified planting material is reflected in the higher oil extraction rates. SIGNIFICANCEIncreased smallholder oil palm yields on existing plantations through improved plant nutrition offers the opportunity to improve smallholder profits and livelihoods, whilst at the same time increasing total oil production without bringing new areas into cultivation.

Agronomy explains large yield gaps in smallholder oil palm fields

CONTEXTPalm oil production is a major source of income for millions of smallholders in Indonesia. However, actual yield remains low in relation to the attainable yield. While emphasis has been given to planting material as a factor explaining low yield in smallholder fields, the role of agronomic management has received less attention. OBJECTIVEWe assessed the role of agronomic management in explaining low oil palm productivity in independent smallholder fields in Indonesia. METHODSWe quantified yield gaps and their causes using data on fresh fruit bunch (FFB) yield and management practices collected over two years across 977 fields. The yield gap was estimated for each field as the difference between actual and attainable yield. The latter was estimated using a crop model, local weather and field-specific soil and palm age. Statistical techniques were used to identify factors explaining yield gaps. RESULTS AND CONCLUSIONSAverage yield represented 42% of the attainable yield, with largest yield gaps occurring in young plantations. Our analyses identified agronomic practices explaining yield gaps, including nutrient management, harvest interval, weed control, and pruning. Improved agronomic management can help smallholders increase yield and profit on existing plantations. SIGNIFICANCEThere is an opportunity to produce more palm oil on current plantation area via better agronomy, generating a positive economic impact at local to national levels, while reducing pressure to clear new land for cultivation. Research & extension programs should be re-oriented to give more emphasis to agronomic management.

The effects of crop type, landscape composition and agroecological practices on biodiversity and ecosystem services in tropical smallholder farms

Abstract In the tropics, smallholder farming characterizes some of the world's most biodiverse landscapes. Agroecology as a pathway to sustainable agriculture has been proposed and implemented in sub‐Saharan Africa, but the effects of agricultural practices in smallholder agriculture on biodiversity and ecosystem services are understudied. Similarly, the contribution of different landscape elements, such as shrubland or grassland cover, on biodiversity and ecosystem services to fields remains unknown. We selected 24 villages situated in landscapes with varying shrubland and grassland cover in Malawi. In each village, we assessed biodiversity of eight taxa and ecosystem services in relation to crop type, shrubland and grassland cover and the number of agroecological pest and soil management practices on smallholder's fields of different crop types (bean monoculture, maize‐bean intercrop and maize monoculture). Increasing shrubland cover altered carabid and soil bacteria communities. Carabid abundance increased in maize but decreased in intercrop and bean fields with increasing shrubland cover. Carabid abundance and richness and wasp abundance increased with soil management practices. Carabid, spider and parasitoid abundances were higher in bean monocultures, but this was modulated by surrounding shrubland cover. Natural enemy abundances in beans were especially high in landscapes with little shrubland, possibly leading to lower bean damage in monocultures compared to intercropped fields, whereas maize monocultures had higher damage. In maize, grassland cover and pest management practices were positively related to damage. Carabid abundance was higher fields with high bean damage, and increased carabid richness in fields with high maize damage. Parasitoid abundance was negatively associated with bean damage. Synthesis and application. Our results suggest that maintaining biodiversity and ecosystem services on smallholder farms is not achievable with a ‘one size fits all’ approach but should instead be adapted to the landscape context and the priorities of smallholders. Shrubland is important to maintain carabid and soil bacterial diversity, but legume cultivation beneficial to natural enemies could complement pest management in landscapes with a low shrubland cover. An increased number of agroecological soil management practices can lead to improved pest control while the effectiveness of agroecological pest management practices needs to be re‐evaluated.

Using Deep Learning and Very-High-Resolution Imagery to Map Smallholder Field Boundaries

The mapping of field boundaries can provide important information for increasing food production and security in agricultural systems across the globe. Remote sensing can provide a viable way to map field boundaries across large geographic extents, yet few studies have used satellite imagery to map boundaries in systems where field sizes are small, heterogeneous, and irregularly shaped. Here we used very-high-resolution WorldView-3 satellite imagery (0.5 m) and a mask region-based convolutional neural network (Mask R-CNN) to delineate smallholder field boundaries in Northeast India. We found that our models had overall moderate accuracy, with average precision values greater than 0.67 and F1 Scores greater than 0.72. We also found that our model performed equally well when applied to another site in India for which no data were used in the calibration step, suggesting that Mask R-CNN may be a generalizable way to map field boundaries at scale. Our results highlight the ability of Mask R-CNN and very-high-resolution imagery to accurately map field boundaries in smallholder systems.

Effect of Time of Tapping on Latex Yield of Rubber (Hevea brasiliensis) in Kalutara and Moneragala Districts of Sri Lanka

Low productivity is one of the major concerns in rubber plantations and therefore, efforts should be made to exploit the maximum potential yield from a rubber tree sustainably. The exploitation of latex (tapping), which is the most important operation in rubber crop and plantations, is normally done in the morning (around 06:00-08:30) in traditional rubber growing areas in Wet and Intermediate Zones and around 05:00-07:30 in the nontraditional areas in Intermediate Zone. Some smallholders, planters and other people/groups believe or suggest that night or early morning tapping could increase the latex yield in rubber trees. Therefore, this study was aimed at exploring the effect of time of taping on the latex yield of H. brasiliensis. Rubber trees of clone RRIC 121 were used with a plot size of 15 trees and six treatments to represent times of tapping [03:00, 04:00, 05:00, 06:00 (control), 07:00 and 08:00)] at d2 frequency at Galewatta division of the Rubber Research Institute, Agalawatta, Kalutara District and 02:00, 03:00, 04:00, 05:00 (control), 06:00 and 07:00 at a smallholder field, Yudaganawa, Moneragala District for one year. Weather and latex volume data were recorded and total crop and g/t/t (gram/tree/tapping) were calculated/estimated. No significant differences were recorded for the mean total crop and g/t/t values for the trees tapped at different time intervals in both Kalutara and Moneragala districts. Models were fit to estimate the crops that could be obtained at different time durations using these data. Accordingly, a 0.04% increase in the total crop was estimated for trees tapped during 5:00-7:30 when compared to those tapped during 06:00-08:30 in the Kalutara District. Conversely, 1.6%, 3.3%, and 5.1% increase in yield were recorded in trees tapped during 04:00- 06:30, 03:00-05:30 and 02:00-04:30 respectively when compared to those tapped during 05:00-07:30. Therefore, night/early morning tapping may not be an effective or a practical strategy/solution to increase the productivity in rubber plantations in Kalutara District, while tapping during 02:00-04:00 in Moneragala District may contribute to an increase in latex crop upto around 5%. Large-scale assessments are required for further confirmation of these results.
 Keywords: Hevea brasiliensis, Latex, Morning, Rubber, Tapping

UNSUPERVISED SEGMENTATION OF SMALLHOLDER FIELDS IN MOZAMBIQUE USING PLANETSCOPE IMAGERY

Abstract. Smallholders produce about a third of the global crop production. Supporting these smallholder farms is an important lever for poverty alleviation. Farm and field sizes are key indicators of many smallholder dynamics, including fragmentation, farm consolidation, and interactions between smallholders, medium-scale commercial farming, and large enterprises. Despite the socio-economic, environmental, and political importance of these dynamics, spatially explicit data on farms and field sizes are still lacking. Identifying small-scale agriculture using satellite imagery is challenging due to the heterogeneity in the crop types and management practices.This study compared three unsupervised segmentation approaches that have not been widely explored for delineating smallholder fields: mean shift, multiresolution segmentation, and simple non-iterative clustering (SNIC), using PlanetScope imagery. The study area is located in northern Mozambique, where 71% of the farms cover less than 2 ha. The results were evaluated using four segmentation accuracy metrics based on object geometries: Area Fit Index (AFI), Quality Rate (QR), Oversegmentation (OS), and Undersegmentation (US). The results showed that the multiresolution segmentation algorithm outperformed the other methods to delineate smallholder fields. This work will support future regional-scale mapping efforts.

Maize Yield Estimation in Intercropped Smallholder Fields Using Satellite Data in Southern Malawi

Satellite data provide high potential for estimating crop yield, which is crucial to understanding determinants of yield gaps and therefore improving food production, particularly in sub-Saharan Africa (SSA) regions. However, accurate assessment of crop yield and its spatial variation is challenging in SSA because of small field sizes, widespread intercropping practices, and inadequate field observations. This study aimed to firstly evaluate the potential of satellite data in estimating maize yield in intercropped smallholder fields and secondly assess how factors such as satellite data spatial and temporal resolution, within-field variability, field size, harvest index and intercropping practices affect model performance. Having collected in situ data (field size, yield, intercrops occurrence, harvest index, and leaf area index), statistical models were developed to predict yield from multisource satellite data (i.e., Sentinel-2 and PlanetScope). Model accuracy and residuals were assessed against the above factors. Among 150 investigated fields, our study found that nearly half were intercropped with legumes, with an average plot size of 0.17 ha. Despite mixed pixels resulting from intercrops, the model based on the Sentinel-2 red-edge vegetation index (VI) could estimate maize yield with moderate accuracy (R2 = 0.51, nRMSE = 19.95%), while higher spatial resolution satellite data (e.g., PlanetScope 3 m) only showed a marginal improvement in performance (R2 = 0.52, nRMSE = 19.95%). Seasonal peak VI values provided better accuracy than seasonal mean/median VI, suggesting peak VI values may capture the signal of the dominant upper maize foliage layer and may be less impacted by understory intercrop effects. Still, intercropping practice reduces model accuracy, as the model residuals are lower in fields with pure maize (1 t/ha) compared to intercropped fields (1.3 t/ha). This study provides a reference for operational maize yield estimation in intercropped smallholder fields, using free satellite data in Southern Malawi. It also highlights the difficulties of estimating yield in intercropped fields using satellite imagery, and stresses the importance of sufficient satellite observations for monitoring intercropping practices in SSA.

High Resolution, Annual Maps of Field Boundaries for Smallholder-Dominated Croplands at National Scales.

Mapping the characteristics of Africa’s smallholder-dominated croplands, including the sizes and numbers of fields, can provide critical insights into food security and a range of other socioeconomic and environmental concerns. However, accurately mapping these systems is difficult because there is 1) a spatial and temporal mismatch between satellite sensors and smallholder fields, and 2) a lack of high-quality labels needed to train and assess machine learning classifiers. We developed an approach designed to address these two problems, and used it to map Ghana’s croplands. To overcome the spatio-temporal mismatch, we converted daily, high resolution imagery into two cloud-free composites (the primary growing season and subsequent dry season) covering the 2018 agricultural year, providing a seasonal contrast that helps to improve classification accuracy. To address the problem of label availability, we created a platform that rigorously assesses and minimizes label error, and used it to iteratively train a Random Forests classifier with active learning, which identifies the most informative training sample based on prediction uncertainty. Minimizing label errors improved model F1 scores by up to 25%. Active learning increased F1 scores by an average of 9.1% between first and last training iterations, and 2.3% more than models trained with randomly selected labels. We used the resulting 3.7 m map of cropland probabilities within a segmentation algorithm to delineate crop field boundaries. Using an independent map reference sample (n = 1,207), we found that the cropland probability and field boundary maps had respective overall accuracies of 88 and 86.7%, user’s accuracies for the cropland class of 61.2 and 78.9%, and producer’s accuracies of 67.3 and 58.2%. An unbiased area estimate calculated from the map reference sample indicates that cropland covers 17.1% (15.4–18.9%) of Ghana. Using the most accurate validation labels to correct for biases in the segmented field boundaries map, we estimated that the average size and total number of field in Ghana are 1.73 ha and 1,662,281, respectively. Our results demonstrate an adaptable and transferable approach for developing annual, country-scale maps of crop field boundaries, with several features that effectively mitigate the errors inherent in remote sensing of smallholder-dominated agriculture.

Model for Predicting Rice Yield from Reflectance Index and Weather Variables in Lowland Rice Fields

Smallholder rice farmers need a multi-purpose model to forecast yield and manage limited resources such as fertiliser, irrigation water supply in-season, thus optimising inputs and increasing rice yield. Active sensing tools like Canopeo and GreenSeeker-NDVI have provided the opportunity to monitor crop health and development at different growth stages. In this study, we assessed the effectiveness of in-season estimation of rice yield in lowland fields of northwest Cambodia using weather data and vegetation cover information measured with; (1) the mobile app-Canopeo, and (2) the conventional GreenSeeker hand-held device that measures the normalised difference vegetative index (NDVI). We collected data from a series of on-farm field experiments in the rice-growing regions in 2018 and 2019. Average temperature and cumulative rainfall were calculated at panicle initiation and pre-heading stages when the crop cover index was measured. A generalised additive model (GAM) was generated using log-transformed data for grain yield, with the combined predictors of canopy cover and weather data during panicle initiation and pre-heading stages. The pre-heading stage was the best stage for grain yield prediction with the Canopeo-derived vegetation index and weather data. Overall, the Canopeo index model explained 65% of the variability in rice yield and Canopeo index, average temperature and cumulative rainfall explained 5, 65 and 56% of the yield variability in rice yield, respectively, at the pre-heading stage. The model (Canopeo index and weather data) evaluation for the training set between the observed and the predicted yield indicated an R2 value of 0.53 and root mean square error (RMSE) was 0.116 kg ha−1 at the pre-heading stage. When the model was tested on a validation set, the R2 value was 0.51 (RMSE = 925.533 kg ha−1) between the observed and the predicted yield. The NDVI-weather model explained 62% of the variability in yield, NDVI, average temperature and cumulative rainfall explained 3, 62 and 54%, respectively, of the variability in yield for the training set. The NDVI-weather model evaluation for the training set showed a slightly lower fit with R2 value of 0.51 (RMSE = 0.119 kg ha−1) between the observed and the predicted yield at pre-heading stage. The accuracy performance of the model indicated an R2 value of 0.46 (RMSE = 979.283 kg ha−1) at the same growth stage for validation set. The vegetation-derived information from Canopeo index-weather data increasingly correlated with rice yield than NDVI-weather data. Therefore, the Canopeo index-weather model is a flexible and effective tool for the prediction of rice yield in smallholder fields and can potentially be used to identify and manage fertiliser and water supply to maximise productivity in rice production systems. Data availability from more field experiments are needed to test the model’s accuracy and improve its robustness.

Effects of storage in triple-layer hermetic bags on stored field pea grain quality and infestation by the pea weevil, Bruchus pisorum L. (Coleoptera: Bruchidae)

Field pea is an important pulse crop that plays a key role in human and animal nutrition as well as for soil fertility in many cropping systems. In East Africa, the pea weevil, Bruchus pisorum L. (Coleoptera: Bruchidae), is one of the main biotic production constraints causing considerable loss both in the field and during storage. The storage loss level depends on the number of weevils remaining in the seed after harvest and on the storage facilities and practices of the farmers. To address the storage losses, we evaluated the effectiveness of hermetic triple-layer Purdue Improved Crop Storage (PICS) bags, on stored grain quality and control of pea weevil infestation in Ethiopia. The PICS bag and two currently used farmers’ storage practices: polypropylene bag (PP) and polyethylene bag with Actellic dust (PE) were compared. Each bag was filled with 50 kg of field peas, and stored for 5 months. Fewer live adult pea weevils and lower percent grain damage was recorded in PICS and PE bags compared to PP bags along the storage period. Field pea grains stored in PICS bags and PE bag also retained higher germination capacity as compared to those stored in a PP bag. The proximate composition differed significantly between storage treatments during the trial. Grain stored in PICS bag and PE bag with Actellic dust showed higher grain moisture, crude protein, and crude fat content, but less total carbohydrate and ash content as compared with grain stored in PP bag. We concluded that PICS bags can be used by smallholder field pea growers to maintain grain quality postharvest, remove the need to use pesticides on their stored peas, and reduce pea weevil damage.