Prasad Daggupati

NDVI-based approaches offer an efficient method for estimating the C-factor, providing continuous spatial coverage and enabling monitoring of short-term changes in vegetation and management practices. This study aims to evaluate the performance of nine well-established NDVI-based C-factor models compared to RUSLE2 model estimates across a specific crop type, different tillage methods, and multiple time scales (monthly, seasonal, and yearly). While some NDVI models showed promising agreement with RUSLE2 estimates, this alignment was not sufficient to ensure accurate C-factor representation in the Gully Creek watershed. The results show that NDVI-based model performance varies systematically with crop type, tillage practice, and temporal scale. Monthly estimates generally reflect broader seasonal patterns, indicating that finer temporal resolution captures intra-seasonal variability without altering overall trends. These findings highlight the importance of accounting for spatial and temporal heterogeneity in C-factor estimation, as model effectiveness depends on local crop composition, management intensity, and temporal resolution rather than a single universally applicable approach.

Soil erosion, particularly ephemeral gully (EG) erosion, poses a significant threat to agricultural sustainability and ecosystem health. Despite their substantial impact on soil degradation, EGs have been relatively understudied, primarily due to their temporary nature and the limitations of existing measurement techniques. This study introduces an integrated approach for quantifying and analyzing EGs, addressing the critical need for accurate and scalable measurement methods. Our methodology combines three key components: (1) an updated portable field tool (Gulliometer), which improves upon existing designs to enhance data collection in diverse field conditions; (2) a standardized image acquisition protocol that ensures consistent, high-quality data capture; and (3) an image processing technique leveraging easy repetitive analysis of gully cross-sections. Laboratory validation using known geometric shapes demonstrated the high precision of our methodology, with error rates below 1%. Field applications in two distinct locations in Ontario, Canada, further confirmed the practicality and effectiveness of our approach under varied environmental conditions. This approach not only advances our understanding of ephemeral gully erosion but also aids in the development of effective soil conservation strategies and informed decision-making in land management.

Understanding and mapping ephemeral gullies (EGs) are vital for enhancing agricultural productivity and achieving food security. This study proposes an upscaling-based strategy to refine the predictive mapping of EGs, utilizing high-resolution Pléiades Neo (0.6 m) and medium-resolution Sentinel-2 (10 m) satellite imagery, alongside ground-truth EGs mapping in Niagara Region, Canada. The research involved generating spectral feature maps using Blue, Green, Red, and Near-infrared spectral bands, complemented by indices indicative of surface wetness, vegetation, color, and soil texture. Employing the Random Forest (RF) algorithm, this study executed three distinct strategies for EGs identification. The first strategy involved direct calibration using Sentinel-2 spectral features for 10 m resolution mapping. The second strategy utilized high-resolution Pléiades Neo data for model calibration, enabling EGs mapping at resolutions of 0.6, 2, 4, 6, and 8 m. The third, or upscaling strategy, applied the high-resolution calibrated model to medium-resolution Sentinel-2 imagery, producing 10 m resolution EGs maps. The accuracy of these maps was evaluated against actual data and compared across strategies. The findings highlight the Variable Importance Measure (VIM) of different spectral features in EGs identification, with normalized near-infrared (Norm NIR) and normalized red reflectance (Norm Red) exhibiting the highest and lowest VIM, respectively. Vegetation-related indices demonstrated a higher VIM compared to surface wetness indices. The overall classification error of the upscaling strategy at spatial resolutions of 0.6, 2, 4, 6, 8, and 10 m (Upscaled), as well as that of the direct Sentinel-2 model, were 7.9%, 8.2%, 9.1%, 10.3%, 11.2%, 12.5%, and 14.5%, respectively. The errors for EGs maps at various resolutions revealed an increase in identification error with higher spatial resolution. However, the upscaling strategy significantly improved the accuracy of EGs identification in medium spatial resolution scenarios. This study not only advances the methodology for EGs mapping but also contributes to the broader field of precision agriculture and environmental management. By providing a scalable and accessible approach to EGs mapping, this research supports enhanced soil conservation practices and sustainable land management, addressing key challenges in agricultural sustainability and environmental stewardship.

To understand if machine learning algorithms could be employed in agricultural landscapes to estimate N₂O emissions, multiple linear regression (MLR), random forest regression (RFR), support vector regression (SVR) and artificial neural network (ANN) algorithms are tested on an agricultural site in Ontario, Canada. Two scenarios, High Input (HI) and Low Input (LI), were used to check the performance of these algorithms’ using R², RMSE, VE, p-factor, r-factor and visual inspection indicators. The HI consisted of discrete measurements of N₂O, rainfall, temperature, fertilizer application dates, soil nitrate, ammonium content and pH values, whereas the LI scenario did not use the latter three. The results indicated that MLR was inapplicable as the data did not satisfy its fundamental assumptions. RFR, SVR and ANN under HI were able to capture 64% (66%), 59% (63%) and 94% (43%) of the variability of emissions within the training (testing) datasets. Subsequently, these models were able to capture 92%, 29% and 75% of high emissions (>10 gm/ha/day) within their predictive intervals of 95% confidence. RFR, SVR and ANN under the LI scenario captured 72% (68%), 61% (66%) and 81% (68%) of the variability in N₂O emissions within the training (testing) datasets. While these models were found to have comparable performance in both HI and LI scenarios, HI was found to be better at capturing high emissions. Based on the computational cost, ease in finetuning, capture of peak emissions and stable model performance, RFR and ANN are recommended to estimate N₂O emissions in the study area and similar agricultural landscapes in future studies.

To reduce the potential threat of soil loss due to ephemeral gullies, it is crucial to adopt Best Management Practices (BMPs) that prevent damage to landscapes by reducing sediments load. This study combines two approaches to evaluate and optimize BMPs for reducing sediment load from sheet/rill and ephemeral gully erosion. The research applied a novel methodology integrating a genetic algorithm with the Annualized Agricultural Non-Point Source Pollution model (AnnAGNPS) to optimize the model and also strategically select and place BMPs in Southern Ontario, Canada, to reduce sediment load cost-effectively. The study assessed five BMPs: cover crops, grassed waterways, no-till, conservation tillage, and riparian buffer strips. Considering the average annual sediment load, riparian buffer strips were consistently successful in decreasing average annual sediment load of sheet/rill erosion, with 69% reduction efficiency. Similarly, grassed waterways were the most effective BMPs for reducing average annual sediment load of ephemeral gully erosion, with an efficiency of 81%. These BMPs were integrated into a cost-optimization framework, demonstrating that strategic placement of BMPs could enhance their efficiency. The optimized placement reduced sheet/rill sediment load by 84.6%, ephemeral gully by 85.4%, and total erosion by 86.3%, achieving these results at minimal cost. The study highlights the significance of targeted BMP placement rather than uniform implementation across entire watersheds. This integrated approach is a viable solution for watersheds with limited resources, facilitating decision-makers and aiding in the adoption of BMPs that can comprehensively reduce sediment load. The developed model in the current study can be applied by decision makers in other watersheds with limited resources for implementing BMPs.

Mapping Ephemeral Gullies (EGs) is essential for enhancing the management and conservation of natural and agricultural resources. This study aimed to devise a novel approach utilizing the Dempster-Shafer (D-S) theory to achieve more accurate EGs mapping. To accomplish this, a high-resolution satellite image and ground data collected during a field visit to the Niagara region in Canada were utilized. Firstly, spectral features that effectively identify EGs were extracted from the satellite imagery. Subsequently, three machine learning classifiers including the artificial neural network (ANN), support vector machine (SVM), and random forest (RF) were employed to generate the EGs map. Finally, the D-S theory was applied to integrate the outcomes from these classifiers, aiming for a more precise EGs map. The results highlighted that the most significant variable importance was attributed to the Normalized Near-Infrared (NIR) (18%) and Soil line (15%). The average producer and user accuracies for the EGs and non-EGs classes across the three classifiers were 0.53 and 0.67, and 0.97 and 0.95, respectively. Incorporating the D-S theory enhanced these accuracy values by 0.15 and 0.19 for EGs and 0.02 and 0.02 for non-EGs. Furthermore, the overall accuracies for the EGs maps generated by the ANN, SVM, RF, and D-S theory models were 94%, 93%, 95%, and 97%, respectively. The results of this study showed that D-S theory is useful for improving the EGs mapping using remote sensing data.