Near real-time soybean phenology detection using proximally sensed hyperspectral canopy reflectance and machine learning methods

International Journal of Remote Sensing, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 24

Published: April 15, 2025

Language: Английский

---

Toward a More Robust Estimation of Forest Biomass Carbon Stock and Carbon Sink in Mountainous Region: A Case Study in Tibet, China

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(9), P. 1481 - 1481

Published: April 23, 2024

Mountainous forests are pivotal in the global carbon cycle, serving as substantial reservoirs and sinks of carbon. However, generating a reliable estimate remains considerable challenge, primarily due to lack representative situ measurements proper methods capable addressing their complex spatial variation. Here, we proposed deep learning-based method that combines Residual convolutional neural networks (ResNet) with measurements, microwave (Sentinel-1 VOD), optical data (Sentinel-2 Landsat) forest biomass track its change over mountainous regions. Our approach, integrating across elevations multi-source remote sensing images, significantly improves accuracy estimation Tibet’s (R2 = 0.80, root mean squared error 15.8 MgC ha−1). Moreover, ResNet, which addresses vanishing gradient problem by introducing skip connections, enables extraction patterns from limited datasets, outperforming traditional optical-based or pixel-based methods. The value was estimated 162.8 ± 21.3 ha−1, notably higher than at comparable latitudes flat regions China. Additionally, our findings revealed sink 3.35 TgC year−1 during 2015–2020, is largely underestimated previous estimates, mainly underestimation stock. significant density, combined regions, emphasizes urgent need reassess mountain better approximate budget.

Language: Английский

---

Algorithms for Plant Monitoring Applications: A Comprehensive Review

Algorithms, Journal Year: 2025, Volume and Issue: 18(2), P. 84 - 84

Published: Feb. 5, 2025

Many sciences exploit algorithms in a large variety of applications. In agronomy, amounts agricultural data are handled by adopting procedures for optimization, clustering, or automatic learning. this particular field, the number scientific papers has significantly increased recent years, triggered scientists using artificial intelligence, comprising deep learning and machine methods bots, to process crop, plant, leaf images. Moreover, many other examples can be found, with different applied plant diseases phenology. This paper reviews publications which have appeared past three analyzing used classifying agronomic aims crops applied. Starting from broad selection 6060 papers, we subsequently refined search, reducing 358 research articles 30 comprehensive reviews. By summarizing advantages applying analyses, propose guide farming practitioners, agronomists, researchers, policymakers regarding best practices, challenges, visions counteract effects climate change, promoting transition towards more sustainable, productive, cost-effective encouraging introduction smart technologies.

Language: Английский

---

Determination of rice (Oryza sativa L.) drought stress levels based on chlorophyll a fluorescence through independent component analysis

Photosynthetica, Journal Year: 2025, Volume and Issue: 63(1), P. 73 - 80

Published: March 27, 2025

Sensing rice drought stress is crucial for agriculture, and chlorophyll a fluorescence (ChlF) often used. However, existing techniques usually rely on defined feature points the OJIP induction curve, which ignores rich physiological information in entire curve. Independent Component Analysis (ICA) can effectively preserve independent features, making it suitable capturing drought-induced changes. This study applies ICA Support Vector Machine (SVM) to classify levels using The results show that 20-dimensional ChlF features obtained by provide superior classification performance, with Accuracy, Precision, Recall, F1-score, Kappa coefficient improving 18.15%, 0.18, 0.17, 0.22, respectively, compared work provides determination method highlights importance of applying dimension reduction methods analysis. expected enhance detection ChlF.

Language: Английский

---

High Spatial Resolution Fractional Vegetation Coverage Inversion Based on UAV and Sentinel-2 Data: A Case Study of Alpine Grassland

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(17), P. 4266 - 4266

Published: Aug. 30, 2023

Fractional vegetation coverage (FVC) is an important indicator of ecosystem change. At present, FVC products are mainly concentrated at low and medium spatial resolution lack high temporal resolution, which brings certain challenges to the fine monitoring ecological environments. In this study, we evaluated accuracy four remote sensing inversion models for based on high-spatial-resolution Sentinel-2 imagery unmanned aerial vehicle (UAV) field-measured data in 2019. Then were optimized by constructing a multidimensional feature dataset. Finally, Source Region Yellow River (SRYR) product was created using best model, spatial-temporal variation characteristics region analyzed. The study’s findings revealed that: (1) accuracies as follows: Gradient Boosting Decision Tree (GBDT) model (R2 = 0.967, RMSE 0.045) > Random Forest (RF) 0.962, 0.049) Support Vector Machine (SVM) 0.925, 0.072) Pixel Dichotomy (PD) 0.869, 0.097). (2) Constructing dataset optimize driving can improve model. NDVI elevation factors affecting machine learning algorithms, visible blue band most factor GBDT (3) SRYR gradually increased from west east north south. change trajectories grassland 2017 2022 not significant. areas that tend distributed southeast (1.31%), while degrade central northwest (1.89%). This study provides optimization scheme, great significance alpine

Language: Английский

---

Hyperspectral Indices Developed from Fractional-Order Derivative Spectra Improved Estimation of Leaf Chlorophyll Fluorescence Parameters

Plants, Journal Year: 2024, Volume and Issue: 13(14), P. 1923 - 1923

Published: July 12, 2024

Chlorophyll fluorescence (ChlF) parameters offer valuable insights into quantifying energy transfer and allocation at the photosystem level. However, tracking their variation based on reflectance spectral information remains challenging for large-scale remote sensing applications ecological modeling. Spectral preprocessing methods, such as fractional-order derivatives (FODs), have been demonstrated to advantages in highlighting features. In this study, we developed assessed ability of novel indices derived from FOD spectra other transformations retrieve ChlF various species leaf groups. The results obtained showed that empirical were low reliability estimating parameters. contrast, low-order a significant improvement estimation. Furthermore, incorporation specificity enhanced non-photochemical quenching (NPQ) sunlit leaves (R2 = 0.61, r 0.79, RMSE 0.15, MAE 0.13), fraction PSII open centers (qL) shaded 0.50, 0.71, 0.09, 0.08), quantum yield (ΦF) 0.85, 0.002, 0.001). Our study demonstrates potential capturing variations Nevertheless, given complexity sensitivity parameters, it is prudent exercise caution when utilizing them.

Language: Английский

---

Dynamic Monitoring and Analysis of Ecological Environment Quality in Arid and Semi-Arid Areas Based on a Modified Remote Sensing Ecological Index (MRSEI): A Case Study of the Qilian Mountain National Nature Reserve

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(18), P. 3530 - 3530

Published: Sept. 23, 2024

The ecosystems within the Qilian Mountain National Nature Reserve (QMNNR) and its surrounding areas have been significantly affected by changes in climate land use, which have, turn, constrained region’s socio-economic development. This study investigates regional characteristics application requirements of ecological environment arid semi-arid zones reserve. In view saturated NDVI reserve high-altitude saline-alkali environmental conditions, this proposed a Modified Remote Sensing Ecology Index (MRSEI) introducing kernel comprehensive salinity index (CSI). approach enhances applicability remote sensing index. temporal spatial dynamics quality QMNNR from 2000 to 2022 were quantitatively assessed using MRSEI. effect use on was quantified analyzing MRSEI contribution rate. findings paper indicate that (1) regions, provides more precise representation surface compared (RSEI). high correlation (R2 = 0.908) significant difference between RSEI demonstrate accuracy evaluating quality. impact rate (2) exhibited an upward trend 2022, with increase 1.3 × 10−3 y−1. area characterized improved constitutes approximately 53.68% total area. Conversely, degraded accounts for roughly 28.77%. (3) Among various types, improvement is primarily attributed expansion forest grassland areas, along reduction unused land. Forest types account over 90% classified “good” “excellent” grades, whereas represent than 44% “poor” grades. Overall, valuable framework regions.

Language: Английский

---

Optimizing natural boundary definition and functional zoning in protected areas: An integrated framework encompassing species, landscapes and ecosystems

Global Ecology and Conservation, Journal Year: 2023, Volume and Issue: 49, P. e02781 - e02781

Published: Dec. 21, 2023

To promote the harmonized development of economic construction and ecological protection, our study introduces an integrated framework that employs various methodologies to delineate natural reserve boundaries spatial zoning. These aim address issues such as insufficient protected area, excessive human-induced influences, inadequate protection endangered animals within nature boundaries. Leveraging comprehensive data from diverse sources, including ground surveys remote sensing detection, we conducted a survey using Chebaling National Nature Reserve in China its environs case study. Models maximum entropy model (MaxEnt), Fragstats, Integrated Valuation Ecosystem Services Trade-offs (InVEST) were employed identify areas with highly suitable habitats, significant landscape diversity, superior ecosystem quality for 16 key species. Subsequently, irreplaceable value research area was calculated Marxan model, leading establishment novel boundary plan. We propose expanding original 1344 km², dividing it into core (321 23.88%) general control (1023 76.12%). Additionally, recommend further division several functional zones facilitate integration diversity protection. This contributes more scientifically informed rational management approach Reserve. Moreover, this offers valuable insights assessing identifying animal habitats globally spatially zoning other reserves.

Language: Английский

---

Application of Multi-Source Remote Sensing Data and Machine Learning for Surface Soil Moisture Mapping in Temperate Forests of Central Japan

Information, Journal Year: 2024, Volume and Issue: 15(8), P. 485 - 485

Published: Aug. 15, 2024

Surface soil moisture (SSM) is a key parameter for land surface hydrological processes. In recent years, satellite remote sensing images have been widely used SSM estimation, and many methods based on satellite-derived spectral indices also to estimate the content in various climatic conditions geographic locations. However, achieving an accurate estimation of at high spatial resolution remains challenge. Therefore, improving precision through synergies multi-source data has become imperative, particularly informing forest management practices. this study, integration with random support vector machine models was conducted using Google Earth Engine order develop maps temperate forests central Japan. The synergy Sentinel-2 terrain factors, such as elevation, slope, aspect, slope steepness, valley depth, model provided most suitable approach yielding highest accuracy values (overall testing = 91.80%, Kappa 87.18%, r 0.98) This finding provides more valuable information mapping, which shows promise forestry applications.

Language: Английский

---

Three Decades of Inundation Dynamics in an Australian Dryland Wetland: An Eco-Hydrological Perspective

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(17), P. 3310 - 3310

Published: Sept. 6, 2024

Wetland ecosystems are experiencing rapid degradation due to human activities, particularly the diversion of natural flows for various purposes, leading significant alterations in wetland hydrology and their ecological functions. However, understanding quantifying these eco-hydrological changes, especially concerning inundation dynamics, presents a formidable challenge lack long-term, observation-based spatiotemporal information. In this study, we classified areas into ten equal-interval classes based on probability derived from dense, 30-year time series Landsat-based maps over an Australian dryland riparian wetland, Macquarie Marshes. These were then compared with three simplified vegetation patches area: river red gum forest, woodland, shrubland. Our findings reveal higher small area covered by exhibiting persistent time. contrast, woodland shrubland show fluctuating patterns. When comparing percentage Normalized Difference Vegetation Index (NDVI), observed notable agreement peaks, lag NDVI response. A strong correlation between inundated was found patch. During dry, wet, intermediate years, patch consistently demonstrated similar probabilities, while exhibited variable probabilities. drying events, dried faster, likely evaporation rates driven exposure solar radiation. long-term SAGA wetness index, highlighting influence topography probability. provide crucial insights complex interactions hydrological processes dynamics ecosystems, underscoring need comprehensive monitoring management strategies mitigate preserve vital ecosystems.

Language: Английский

---