BEGIN:VCALENDAR VERSION:2.0 PRODID:-//pretalx//talks.osgeo.org//foss4g-2026//talk//JMNGAW BEGIN:VTIMEZONE TZID:JST BEGIN:STANDARD DTSTART:20000101T000000 RRULE:FREQ=YEARLY;BYMONTH=1 TZNAME:JST TZOFFSETFROM:+0900 TZOFFSETTO:+0900 END:STANDARD END:VTIMEZONE BEGIN:VEVENT UID:pretalx-foss4g-2026-JMNGAW@talks.osgeo.org DTSTART;TZID=JST:20260902T110000 DTEND;TZID=JST:20260902T113000 DESCRIPTION:Maize is the primary staple crop in Malawi and is predominantly cultivated by smallholder farmers operating on plots typically smaller th an one hectare. Accurate and timely yield estimation is essential for nati onal food security planning\, early warning systems\, and resource allocat ion. However\, traditional field-based yield assessment methods are labour -intensive\, time-consuming\, costly\, and difficult to scale across dispe rsed rural farming systems. With the increasing availability of open-sourc e geospatial tools and freely accessible satellite imagery\, remote sensin g offers a scalable and cost-effective alternative for crop yield monitori ng in low-resource settings.\nThis study evaluates and compares the perfor mance of Sentinel-2\, Landsat 9\, and Unmanned Aerial Vehicle (UAV) imager y for maize yield prediction in selected smallholder fields in Zomba Distr ict\, Malawi. The objective is to assess prediction accuracy across sensor s and demonstrate how open-source geospatial technologies can support scal able yield monitoring in Sub-Saharan Africa.\nThe study area consisted of maize fields ranging from 0.06 to 0.89 hectares\, representative of Malawi ’s smallholder farming systems. Ground-truth yield data were collected a t the end of the 2024–2025 growing season through manual harvesting and weighing of maize cobs. Yields were standardized to tons per hectare to en able comparison across fields of varying sizes.\nSatellite imagery from Se ntinel-2 was accessed through Google Earth Engine and acquired monthly fro m November to March. Landsat 9 imagery was downloaded from the USGS Earth Explorer for the same period\, excluding January due to persistent cloud c over. UAV data were collected in March 2025 using a DJI Matrice 300 RTK fl own at an altitude of 120 meters\, achieving a ground sampling distance of 3.67 cm per pixel. The UAV was equipped with a Zenmuse H20N camera. Drone imagery was processed into orthomosaics using Pix4DMapper before further analysis in QGIS.\nImage processing and feature extraction were conducted using QGIS\, emphasizing the use of open-source tools to ensure accessibil ity and reproducibility. For satellite imagery\, vegetation indices includ ing NDVI\, SAVI\, and GNDVI were computed\, along with extraction of raw R ed\, Green\, and Blue spectral bands. Monthly averages were calculated at the field level. To increase the number of training samples\, each monthly satellite image was treated as an independent observation. For UAV imager y\, high-resolution orthomosaics enabled extraction of RGB-based vegetatio n indices such as Excess Green (ExG)\, Green Leaf Index (GLI)\, Visible At mospherically Resistant Index (VARI)\, Triangular Greenness Index (TGI)\, and Colour Index of Vegetation Extraction (CIVE)\, along with mean RGB val ues. Although UAV imagery was collected only once during crop maturity\, i ts high spatial resolution provided detailed canopy information.\nSeparate linear regression models were developed for each sensor type using extrac ted indices and band features as predictors. Linear regression was selecte d due to its simplicity\, interpretability\, and suitability for small dat asets\, particularly in contexts where field sizes are small and the numbe r of usable pixels\, especially for coarse-resolution imagery such as Land sat 9 with 30-meter spatial resolution\, is limited. More complex machine learning models risk overfitting and may not generalize well in smallholde r farming environments. Model performance was evaluated using the Coeffici ent of Determination (R²)\, Root Mean Squared Error (RMSE)\, and Mean Abs olute Error (MAE). For UAV data\, cross-validation was conducted by rotati ng through fields to ensure robust evaluation.\nResults revealed substanti al differences in predictive performance across sensors. UAV imagery achie ved the highest accuracy\, with an R² of 0.9991\, RMSE of 0.0135 t/ha\, a nd MAE of 0.0109 t/ha\, indicating an almost perfect fit between predicted and observed yields. Sentinel-2 demonstrated moderate predictive capabili ty\, achieving an R² of 0.4063\, RMSE of 0.3537 t/ha\, and MAE of 0.2942 t/ha. Landsat 9 exhibited the lowest performance\, with an R² of 0.2331\, RMSE of 0.4020 t/ha\, and MAE of 0.3365 t/ha. These findings suggest that while UAV imagery provides highly precise yield estimates due to its fine spatial resolution\, satellite-based approaches\, particularly Sentinel-2 \, offer promising scalability despite lower predictive strength in this d ataset.\nThe superior performance of UAV imagery can be attributed to its high spatial resolution\, which captures within-field variability more eff ectively than medium-resolution satellite data. However\, UAV deployment i nvolves higher operational costs\, logistical coordination\, and technical expertise\, which limits its scalability for nationwide monitoring. In co ntrast\, Sentinel-2 and Landsat 9 imagery are freely available\, offer bro ad spatial coverage\, and support continuous time-series analysis\, making them more practical for large-scale agricultural monitoring in low-resour ce environments. Improved performance may be achieved by expanding trainin g datasets\, incorporating longer time series\, or integrating multi-senso r data fusion approaches.\nA key contribution of this study is its demonst ration of a fully open and reproducible workflow using QGIS and freely ava ilable imagery. By leveraging open-source software and public satellite da ta\, the framework reduces dependency on proprietary systems and enables l ocal researchers\, agricultural officers\, and non-governmental organizati ons to implement crop monitoring systems without costly infrastructure. Th is approach strengthens local capacity for geospatial analysis and support s sustainable agricultural monitoring initiatives in Sub-Saharan Africa.\n This research addresses a critical gap in the literature\, as no prior stu dy has systematically compared UAV\, Sentinel-2\, and Landsat imagery for maize yield prediction within Malawi’s smallholder farming context. The findings highlight the trade-offs between spatial resolution\, cost\, scal ability\, and predictive accuracy. While UAV imagery achieved the highest precision\, Sentinel-2 demonstrated potential as a scalable alternative fo r early yield estimation when combined with appropriate modelling approach es and larger datasets.\nIn conclusion\, the integration of open data\, si mple and interpretable models\, and open-source geospatial tools offers a viable pathway for affordable and repeatable crop yield estimation in smal lholder systems. With further refinement and expanded datasets\, satellite -based approaches can play a significant role in supporting food security decision-making across Malawi and similar Sub-Saharan African contexts. Th e methods and processing workflows developed in this study are openly shar ed to facilitate reuse and adaptation by researchers and practitioners aim ing to strengthen agricultural monitoring and planning efforts. DTSTAMP:20260717T225756Z LOCATION:Cosmos1 SUMMARY:Comparative Analysis of Multi-Sensor Responses for Maize Yield Esti mation - Esnart Sandikonda URL:https://talks.osgeo.org/foss4g-2026/talk/JMNGAW/ END:VEVENT END:VCALENDAR