Intercropping legumes with cereals has been a common cropping system in short-season rainfed environments due to its increased productivity and sustainability. Intercropping barley (Hordeum vulgare L.) with Egyptian clover (Trifolium alexandrinum L.) could increase the grain yield of barley and improve resource use efficiency of the intercropping system. However, non-optimum planting geometry has been a hurdle in the adaptation of barley-based cropping systems. This study was aimed at optimizing the planting geometry, and assess the productivity and profitability of barley-Egyptian clover intercropping system. Ten different planting geometries, differing in number of rows of barley, width and number of irrigation furrows and planting method were tested. Intercropping barley with Egyptian clover improved 56–68% grain yield of barley compared with mono-cropped barley. Barley remained dominant crop in terms of aggressiveness, relative crowding coefficient and competitive ratio. The amount of water used was linearly increased with increasing size of barley strip from 3 to 8 rows. The highest water use efficiency (4.83 kg/cf3) was recorded for 8-row barley strip system with 120 cm irrigation furrows compared to rest of the planting geometries. In conclusion, 8-rows of barley planted on beds with Egyptian clover in 120 cm irrigation furrows had the highest net income and cost benefit ratio. Therefore, it is recommended that this planting geometry can be used for better economic returns of barley-Egyptian clover intercropping system. However, barley strips with >8 rows were not included in this study, which is limitation of the current study. Therefore, future studies with >8 barley rows in strip should be conducted to infer the economic feasibility and profitability of wider barley strips.
9 March 2022 in Remote Sensing Applications: Society and Environment
Accurate geospatial inventories of fruit orchards are essential for precision horticulture and food security, yet Pakistan lacks consistent spatial datasets at district and tehsil levels. This study presents the first national-scale, object-based Random Forest (RF) framework for orchard delineation and yield estimation by integrating multi-temporal Sentinel-2 imagery on Google Earth Engine (GEE) with high-resolution Pakistan Remote Sensing Satellite-1 (PRSS-1) data. Among the tested classifiers, RF achieved the highest performance on Sentinel-2 data (Overall Accuracy (OA) = 79.0%, kappa (κ) = 0.78), outperforming Support Vector Machines (OA = 74.5%, κ = 0.74) and Gradient Boosting Decision Trees (OA = 73.8%, κ = 0.73), with statistical significance confirmed (McNemar’s χ2, p < 0.01). Integrating RF with Object-Based Image Analysis (OBIA) on PRSS-1 imagery further enhanced boundary precision (OA = 92.6%, κ = 0.89), increasing Producer’s and User’s accuracies to 90.4% and 91.5%, and increasing Intersection-over-Union (IoU) from 0.71 to 0.86 (p < 0.01). Regression-based yield modeling using field-observed data revealed that mean- and median vegetation index aggregations provided the most stable predictions (R2 = 0.77–0.79; RMSE = 72–105 kg tree−1), while extreme-value models showed higher errors (R2 = 0.46–0.56; RMSE > 560 kg tree−1). The resulting multisensory geospatial inventory of citrus and mango orchards establishes a scalable, transferable, and operationally viable framework for orchard mapping yield forecasting, and resource planning, demonstrating the strategic value of national satellite assets for food security monitoring in data-scarce regions.
Accurate geospatial inventories of fruit orchards are essential for precision horticulture and food security, yet Pakistan lacks consistent datasets at district and tehsil levels. This study develops the first national-scale, ML-enabled framework for orchard delineation and yield estimation by integrating multi-temporal Sentinel-2 imagery on Google Earth Engine (GEE) with high-resolution Pakistan Remote Sensing Satellite-1(PRSS-1) data, advanced machine learning, and object-based image analysis (OBIA). Among tested classifiers, Random Forest (RF) achieved the highest performance on pixel-based Senti-nel-2 data (OA = 79.0%, κ = 0.78) outperforming Support Vector Machines (74.5%, κ = 0.74) and Gradient Boosting Decision Trees (73.8%, κ = 0.73), with improvements confirmed by McNemar’s test (p < 0.01). Integrating RF with OBIA on PRSS-1 imagery further enhanced delineation, increasing OA to 92.6% (κ = 0.89), with producer’s and user’s accuracies of 90.4% and 91.5%, and IoU improving from 0.71 to 0.86 (p < 0.01) in orchards. Yield modeling based on field-observed data showed mean- and median-based vegetation index aggregation performed best (R² = 0.77–0.79; error = 72–105 kg/tree), while extreme-value models performed poorly (R² = 0.46–0.56; error >560 kg/tree). The resulting validated geospatial inventory of citrus (Central Punjab) and mango (South Punjab) supports productivity forecasting, export planning, and resource allocation, demonstrating the transformative role of national space assets in advancing precision horticulture and strengthening food security in data-scarce regions