Reliable and intelligent retrieval of leaf traits from hyperspectral reflectance is crucial for assessing ecosystem functions, yet conventional approaches struggle with spectral complexity and nonlinearities. To address these challenges, we developed the Leaf Trait Retrieval Network (LTRN), a novel deep learning framework that integrates Kolmogorov-Arnold Network (KAN), Transformer, and Temporal Convolutional Networks (TCN) for end-to-end trait estimation. Model validation was carried out using a large spectral-trait database covering hundreds of plant species and four functional traits. Experimental results demonstrated that LTRN model outperforms state-of-the-art deep learning models, achieving R2 values greater than 0.78 for estimating chlorophyll content (Chla+b), equivalent water thickness (EWT), carotenoid content (Ccar), and leaf mass per area (LMA). Further analyses indicated that the LTRN model delivers stable estimation performance across spectral resolutions of 10-25 nm. Moreover, the model demonstrates strong stability across varying proportions of training samples. These findings underscore the robustness and stability of LTRN for large-scale vegetation trait retrieval, offering a valuable framework for advancing the intelligent estimation of other ecological parameters.

Soil erosion represents a critical threat to natural resource sustainability in arid environments, yet validation of empirical models in hyper-arid watersheds remains limited. This study employs the Revised Universal Soil Loss Equation (RUSLE) integrated with Geographic Information Systems (GIS) and remote sensing to quantify rainfall erosivity and erosion risk dynamics in Wadi Allith basin, Saudi Arabia, across 2016–2018. RUSLE factors, rainfall erosivity (R), soil erodibility (K), slope length-steepness (LS), cover management (C), and support practices (P) were derived from WorldClim precipitation data, FAO-UN Soil Map, SRTM digital elevation model, and Landsat-derived NDVI. Results reveal a pronounced orographic gradient in rainfall erosivity, increasing 29 % from 2016 (120 mm yr−1) to 2018 (155 mm yr−1), with high-erosivity zones expanding from 15 % to 35 % of basin area. Despite this intensification, the basin maintains predominantly “slight” erosion risk (ERC 1–2, <5 t ha−1·yr−1) across 95–97 % of its area, attributed to low mean LS-factor (≈2.3). However, a statistically significant 6 % relative increase in soil loss (0.3 t ha−1·yr−1) occurred, concentrated in 3.2 % of highland tributaries where LS > 10. This demonstrates that while low-gradient topography provides a geomorphic buffer, climate-driven erosivity increases threaten to breach critical erosion thresholds. The successful application of open-source GIS tools confirms cost-effective scalability for long-term monitoring in data-scarce arid regions. These findings inform targeted conservation strategies and highlight the urgency of climate-adaptive soil management in the Arabian Peninsula’s vulnerable dryland watersheds.

Spectral inconsistency is a major obstacle that hinders the widespread application of UV-Visible spectrometry in inland waters. The variance of water spectrum is primarily concentrated in the UV band, leading the existing methods to ignore spectrum features in the visible band and fail to adequately restore subtle spectral details. In addition, existing methods usually rely on a large number of standard samples, making calibration transfer laborintensive and impractical for dynamic field conditions. To address these challenges, we proposed a multi-stage calibration transfer method with minimal standard samples, specifically designed for UV-Visible spectrometry in inland waters. To reduce the number of standard samples and maintain the effectiveness of calibration transfer, Bayesian optimization method was designed to jointly adapt both sample selection and spectral mapping, maximizing the utility of minimal standard samples. Unlike other widely adopted methods that typically require 20 - 60 samples, the proposed approach achieved superior spectral consistency with only 11-13 samples, with mean coefficient of determination (R2) of 0.99 and mean spectral angle of 2.61 degrees. Water quality retrieval models trained on benchtop spectrometers could be directly applied to buoy or handheld spectrometers without additional adjustments, achieving favorable performance (mean R2 = 0.62) comparable to that obtained with benchtop spectrometer (mean R2 = 0.71). To facilitate the field investigation, reference spectrometer can also be switched from benchtop spectrometer to the handheld one. The proposed method significantly improves between-device spectral consistency and facilitates model reuse, which is powerful tool to accelerate application of UV-Visible spectrometry of inland waters, such as buoy-based monitoring and mobile monitoring.

This study develops a novel quantitative risk assessment framework to evaluate the environmental impact of Olive Mill Wastewater (OMW) pollution on surface water resources in the Keritis watershed, Western Crete. Unlike prior OMW assessments that relied on qualitative indicators or single-factor analysis, this research integrates Multi Criteria Analysis (MCA) with physically based hydrological modeling to enable quantitative, sub-catchment scale risk mapping. Risk is assessed through two core components: magnitude (encompassing spatial and temporal pollution dimensions) and probability (integrating hazard occurrence, receptor exposure, and harm likelihood). Eleven criteria, including population vulnerability, precipitation, pollutant dilution potential, and lagoon conditions, were normalized and weighted via the Analytic Hierarchy Process (AHP). The methodology classified sub-catchments into five risk tiers using natural breaks, revealing significant spatial variability. Key findings identified sub-catchments 4, 5, and 9 as high-to-moderate risk zones for human health, while sub-catchments 5 and 6 faced elevated risks for NATURA sites. The analysis underscores the critical role of flow path length and dilution capacity in mitigating risks. Specific evidence-based management strategies are proposed to support EU Water Framework Directive alignment, balancing ecological preservation with socio-economic needs in olive oil-dependent regions.

Abundant hyperspectral remote sensing data provides change detection (CD) technology with the opportunity to accurately distinguish land cover transformations across a consistent area over time. However, most deep learning methods for hyperspectral CD are performed in a fully supervised manner, which requires a large number of pixel-level labeled samples and is highly time-consuming. Therefore, this study combines semi-supervised learning (SSL) to reduce label acquisition costs. However, there are many challenging areas, such as low contrast regions, which increase the extra burden to interpret general semantic information from land cover attributes. To address these challenges, we introduce an innovative multiview prototype learning (MVPL)-based general semantic explaining temporal-spatial-spectral transformer (SETSST) for hyperspectral CD: 1) first, we design self-view and cross-view prototypical similarity maps according to distinct perturbed views of input samples, which ensures appropriate feature interactions by leveraging the low-density aggregated representation; 2) second, the semi-supervised pretraining framework based on weak-to-strong prediction alignment and MVPL is designed to simultaneously capture features from both labeled and unlabeled samples; and 3) to comprehensively interpret semantic explaining information in real land cover attributes, we propose the semantic explaining change analysis (SECA) module. This module enhances feature explanation by clustering them into several change types and an unchanged type, while accounting for the influence of distance to the central pixel. Extensive experiments conducted on three real datasets demonstrate the effectiveness of the proposed MVPL-based semi-supervised pretraining framework and the SECA module in leveraging both labeled and unlabeled samples for hyperspectral CD. This method surpasses the performance of many existing hyperspectral CD approaches and significantly reduces the need for labeled training samples.

Crop classification is critical for precision agriculture, food security, and policymaking. While the fusion of multisensor data and deep learning holds significant potential for enhancing crop classification, the application of deep learning-based spectral matching techniques (SMTs) in this domain remains underexplored. To address this gap, this study proposes a novel time-aware siamese network (TASN) for spectrotemporal signature (STS) matching, focusing on Kharif crops across three districts in Punjab, Pakistan. The research integrates fusion of Landsat-8/9 and Sentinel-2 data to generate a high-spatial and high-temporal (HSHT) dataset, determines optimal temporal lengths for STS extraction using Jeffries-Matusita (JM) distance analysis, applies smoothing techniques to reduce noise in STS, and evaluates TASN's transfer learning capability for cross-regional crop classification. Results demonstrate that the HSHT dataset significantly enhances temporal resolution, enabling precise crop monitoring. The TASN model achieves 95% accuracy with the universal normalized vegetation index (UNVI), 92% with the normalized difference vegetation index (NDVI), and 91% with the enhanced vegetation index (EVI). Transfer learning experiments reveal robust adaptability; applying STS of one region to others, TASN attains 98% accuracy for rice classification in Hafizabad using UNVI and 81%-87% accuracy for multicrop classification in Lodhran. These findings highlight the efficacy of spectral matching for scalable, transferable crop classification. The study advances agricultural remote sensing by introducing a deep learning framework for STS matching and demonstrating its utility across diverse regions. By combining multisensor fusion and TASN's transferability, this study offers a practical tool for policymakers and farmers, supporting scalable crop monitoring and food security initiatives.

Existing algorithms for chlorophyll-a (Chla) retrieval encounter two key challenges: poor cross-domain generalization, and the scarcity of paired field Chla measurements with corresponding spectrum. Water optical properties vary strongly with environmental conditions, causing pronounced distribution shift across different spatial and temporal domains. However, most models assume that spectra in new domains follow the same distribution as the training data, which leads to sharp performance degradation under distribution shifts. Furthermore, the absence of effective sample selection strategy increases data demand for model development, exacerbating conflicts with limited field observations. To address these issues, we proposed FSResTL-Chla, a fewshot ResNet transfer learning method for Chla retrieval. We constructed a universal Chla-spectrum model with a pre-trained ResNet under different optical water types. To ensure the representativeness and informativeness of few-shot samples, a self-adaptive representative sample selection strategy was proposed, coupled with finetuning to adjust the universal model to distribution shift. FSResTL-Chla showed favorable performance across multiple optical water types using 7-12 samples, with an average coefficient of determination (R2) of 0.68. Despite using only 18% of all observations, FSResTL-Chla achieved performance (R2 = 0.52-0.79) comparable to or exceeding that of the best traditional models (R2 = 0.49-0.80), which commonly required 70% - 80% of the observations for model development. This demonstrates that FSResTL-Chla maintains strong spatiotempral generalization, whereas traditional models often fail under changing conditions. By improving cross-domain generalization while reducing dependence on sample size, FSResTL-Chla provides a valuable tool for environmental monitoring and management.

Hyperspectral-multispectral image fusion (HMF) aims to recover a high-resolution hyperspectral image (HrHSI) from an observed high-resolution multispectral image and a low-resolution hyperspectral image. Traditional HMF methods often suffer from over-reliance on degradation priors. While current blind HMF methods reduce this dependence by estimating the spectral degradation process, they still struggle with common real-world challenges, particularly spatial misregistration and spectral range nonoverlap between the observed pair. To address these limitations, we propose a novel blind HMF framework that explicitly tackles both issues. Specifically, we introduce a dedicated image registration strategy before the fusion stage, which explicitly strengthens the spatial mapping relationship between the images and thus eliminates the need for strict initial spatial alignment assumptions. Building upon this, we establish an optimization model for spectral response function (SRF) estimation. The proposed method then employs a decoupled reconstruction mechanism that partitions the HMF process based on spectral band overlap, followed by a spectral recovery module (SRM) to further refine the fused HrHSI, ensuring high spectral accuracy across the entire spectrum. Experimental results demonstrate the exceptional performance of the proposed method in challenging HMF scenarios with significant spatial misregistration. Compared to existing approaches, the proposed method exhibits conclusive superior robustness to spatial misregistration in both SRF estimation and image fusion, achieving higher spectral accuracy in nonoverlapping bands and delivering better spatial-spectral reconstruction. Notably, the SRM in the proposed method is highly portable and can be integrated as a general optimization component into other HMF methods to further enhance their performance.

Accurate monitoring of nitrogen nutrition is critical for optimizing cotton production. Traditional machine learning-based inversion models have limited effectiveness for precision monitoring. Multisource fusion models for small samples were developed in this study to achieve enhanced accuracy through fitting and data complementarity. Cotton plants subjected to different nitrogen treatments were investigated. A two-year pot experiment was conducted to collect main-stem leaf images to construct an image pretraining dataset for model transfer. In a field experiment conducted over one year, main-stem leaf data were collected using hyperspectral, chlorophyll fluorescence, and digital camera sources, thereby providing a multisource dataset for training monitoring models. Two architectures—a neural network (NN) and an interpretable deep forest (DF), which are suitable for small-sample spectral, fluorescence, image color, and texture-sequence features—were constructed to improve the accuracy of nitrogen content inversion. Additionally, a two-dimensional sliding-window processing method was introduced into the DF multigranularity scanning module, and a transfer-learning-based two-dimensional convolutional NN was employed to directly model small-sample two-dimensional images. Building upon the outcome, multilayer fusion models were constructed, with corresponding fusion strategies designed for homogeneous sequence inputs and heterogeneous image–sequence inputs. The results showed that NN and DF can effectively handle limited sample sizes and outperform traditional machine learning models. Among the fusion models, the optimal secondary decision-level fusion model achieved an R² of 0.926 on the independent test set, indicating good performance under small-sample conditions. This study provides a methodological reference for the precise monitoring of crop phenotypic parameters under small-sample conditions.

Navigating complex human-nature interactions remains a critical challenge for protected areas globally. China’s establishment of the National Park System represents a fundamental governance shift toward ecological integrity, yet its long-term efficacy in economically vibrant regions remains underexplored. Focusing on Wuyishan National Park, we reconstructed forest disturbance dynamics from 1991 to 2023 using Landsat time series and the LandTrendr algorithm. Over this period, an estimated 6.0% of the park area (approximately 7,739.4 ha) experienced disturbances. Quantitative analysis revealed that the majority (58.8%) of disturbed forests eventually recovered, whereas tea plantation expansion (23.0%) exerted the primary permanent anthropogenic pressure, clustering selectively in non-protected Gaps and along administrative boundaries, while the Core Protected Area remained safeguarded. Disturbance trajectories aligned with policy evolution, economic incentives, and climatic extremes. Following the 2015 National Park Pilot, disturbances stabilized at minimal levels, coinciding with the effective containment of historical tea expansion. This study establishes a transferable remote sensing framework for evaluating conservation outcomes, highlighting the utility of continuous spatial monitoring in assessing protected areas confronting complex socio-economic and climatic pressures.

Nitrogen, an essential nutrient for cotton growth, requires accurate and timely assessment for effective fertilizer application. Despite advances in sensing technology and models, traditional machine learning monitoring models exhibit limited accuracy in assessing nitrogen content. Agronomic sample collection is challenging, and small datasets are unsuitable for conventional deep learning (DL) methods. Moreover, monitoring at specific time points cannot capture dynamic nitrogen changes within the crop, and time lags in decision-making can lead to mismatches between nitrogen supply and crop demand. Therefore, improving monitoring accuracy with small agronomic samples and effectively predicting future nitrogen content changes are crucial. Here, we used a hyperspectral technology to collect data, focusing on "Xinluzao53" cotton and established four nitrogen concentration gradients. Approximately 30 days after emergence, we conducted destructive and non-destructive sampling of the main stem leaves at regular intervals. To train the monitoring model, destructive sampling involved collecting hyperspectral data, followed by leaf cutting for nitrogen determination, whereas nondestructive sampling involved collecting hyperspectral data over time without leaf damage. We then constructed DL monitoring models suitable for small samples to estimate nitrogen levels. The optimal monitoring model was applied to non-destructive sampling, and the resulting nitrogen content time-series was cleaned and used as input for prediction models. The one-dimensional convolutional neural network monitoring model developed in this study achieved optimal accuracy, and the improved ensemble time-series prediction models demonstrated better predictive performance than single time-series models. These findings offer valuable insights for monitoring phenotypic parameters with limited sample sizes and predicting future changes.

This study develops a quantitative risk assessment framework to evaluate the environmental impact of Olive Mill Wastewater (OMW) pollution on surface water resources in the Keritis watershed, Western Crete. Utilizing a Multi-Criteria Analysis (MCA) approach integrated with hydrological modeling, the research assesses risk through magnitude (spatial and temporal dimensions) and probability (hazard occurrence, receptor exposure, and harm likelihood) components. Eleven criteria, including population vulnerability, precipitation, pollutant dilution potential, and lagoon conditions, were normalized and weighted via the Analytic Hierarchy Process (AHP). The methodology classified sub-catchments into five risk tiers using natural breaks, revealing significant spatial variability. Key findings identified sub-catchments 4, 5, and 9 as high-to-moderate risk zones for human health due to proximity to low-order streams, high phenol concentrations, and precipitation patterns. For NATURA sites, sub-catchments 5 and 6 faced elevated risks, driven by extensive pollutant pathways and habitat sensitivity. The analysis underscores the critical role of flow path length and dilution capacity in mitigating risks, particularly in areas with 4th- and 5th-order streams. Policy implications advocate for restricting olive mill permits in high-risk zones and adopting inorganic flocculation for cost-effective pollution control. The study highlights the framework’s adaptability to diverse pollution scenarios but calls for expanded criteria to address groundwater and soil impacts. By providing a replicable, data-driven tool, this work aids policymakers in prioritizing mitigation efforts and aligning with EU Water Framework Directive goals, balancing ecological preservation with socio-economic needs in olive oil-dependent regions.

The interacting multiple model (IMM) filter algorithm is highly effective in addressing the issue of rapid target maneuvers within tracking scenarios. Nonetheless, owing to the unpredictable nature of target movements and the intricate nature of the measurement system, substantial levels of heavy-tailed process and measurement noises are induced. In the context of the jump Markov system (JMS), we introduce a two-layer feedback adaptive IMM filter (TFAIMM) approach to tackle the instability issues in IMM algorithms, which are influenced by heavy-tailed process and measurement noises. Firstly, the initial fused states are pre-filtered with the involvement of maneuverability. Secondly, the system’s state, degrees of freedom, scale matrix, and target maneuverability parameter are deduced from the pre-filtering outcomes, employing Student’s t-distribution traits and the variational Bayesian (VB) approach. The target maneuverability parameter feedback acts on models probability calculation and pre-filtering. To validate the effectiveness of the TFAIMM in tracking multi-model maneuvering targets, experiments are carried out under various heavy-tailed noise conditions and UWB dataset. Our method is significantly effective in tracking complex targets with uncertain noise.

Long term remote sensing data contains information from four dimensions:time,space,and spectrum.Currently,the description is only based on spatial and spectral dimensions,and there is no unified concept to describe long-term remote sensing data.In traditional research,using a three-dimensional cube model to store long-term remote sensing data would separately store data from different time periods,which cannot meet the efficient storage,fast retrieval,and deep analysis of its time dimension,and thus cannot fully explore the characteristics of different land features changing over time.Essentially,it is still discrete three-dimensional data,and the four-dimensional information of long-term data has not been effectively managed and analyzed uniformly.Therefore,based on the research foundation of predecessors,the concept of time spectrum has emerged,bringing new breakthroughs to the organization and storage of remote sensing data,and improving the utilization value of long-term remote sensing data.Similar to the concept of spectra,remote sensing feature sequences at different times constitute spectrograms. Remote sensing spectrogram theory mainly describes the information of remote sensing data in spectral,temporal,and spatial dimensions.By analyzing the temporal spectrum of remote sensing images,the changing characteristics of surface objects at different time scales and spectral ranges can be revealed,which can be used in various application fields.Experts and scholars in the remote sensing field have fully recognized the scientific value and broad application prospects of MDD. Land change detection and fine classification of crops require the use of multiple spectral information from different time periods.Unlike traditional 3D datasets,the study of remote sensing data has been elevated from 3D to 4D,providing a different way for land features to be represented.This helps to solve the problem of"same spectrum foreign objects,same object different spectrum".The analysis of cultivated land types based on spatiotemporal spectra also has the characteristics of complete phenology.It can effectively eliminate false changes caused by seasonal factors.The MDD format has significantly improved the efficiency and accuracy of remote sensing analysis,and has been widely promoted and applied in multiple industry user units,generating good social and economic benefits. This article summarizes and generalizes the basic concepts,main methods,and techniques of remote sensing spectrogram theory,introduces the application fields and latest research progress of remote sensing spectrogram theory,and finally looks forward to the future research directions of spectrogram theory.The future research directions in this field include the study of spatiotemporal consistency and data fusion of multi-source data,the research of deep learning time-frequency analysis methods,and the research of real-time and high-resolution analysis. In the field of next-generation remote sensing data storage,the multidimensional data format developed by Chinese scientists is expected to meet the needs of real-time online data processing,be used to develop a new spatiotemporal spectral integrated data organization and storage mode,and serve as one of the technical foundations of remote sensing cloud platforms.This innovation can improve the storage,retrieval,and sharing efficiency of remote sensing big data,providing key support for China to build an independent and controllable remote sensing cloud platform.

Monitoring of non-point source pollution (NPSP) in small watersheds suffers from low monitoring frequency and sparse spatial coverage, limiting both our understanding and effective management of NPSP. UV-Visible (UV-Vis) spectrometry, known for its rapid response and cost-effectiveness, offers a promising solution to address this monitoring scarcity. However, key water quality parameters of NPSP, total phosphorus (TP) and total nitrogen (TN) (e.g., ammonia), are inherently challenging to retrieve from UV-Vis spectrum, because TP is optically-nonactive parameter and TN contains optically-nonactive parameter (e.g., ammonia). Moreover, uncertainty evaluation of retrieved water quality relies heavily on concurrent field measurements, lacking effective method for real-time uncertainty assessment. To address these challenges, we constructed a UV-Vis spectrometer network comprising 20 spectrometers and implemented a four-month in-situ NPSP monitoring in Pizhou City, China. Multivariate mixture density network (multi-MDN) was developed to model the covariation relationship among NPSP-related water quality conditioned on the given spectrum, enabling the joint retrieval of TN, TP, and CODMn (chemical oxygen demand). By leveraging this covariation relationship among NPSP-related water quality, the proposed multi-MDN effectively improved generalization capability for both optically-active and optically-nonactive constituents, with R2 of 0.92, 0.84 and 0.90 for TN, TP and CODMn. We further proposed per-estimation uncertainty evaluation method for real-time assessment of NPSP monitoring and designed several experiments to examine its utility. Per-estimation uncertainty effectively identified large retrieval error, detected spectrometer anomalies (e.g., insufficient probe submergence, biofouling), and implied model update. Using the real-time in-situ observation from UV-Visible spectrometer network, we revealed the spatiotemporal variation of NPSP in rural and urban rivers of Pizhou. This study provides an accurate, robust and cost-effective solution to improve spatiotemporal coverage of NPSP monitoring, supporting informed management in small watersheds.

Disease severity grading is a key prerequisite and major aspect of the integrated management of cotton Verticillium wilt (VW). However, the application of current VW severity grading methods requires an investigation into the disease status of all cotton leaves. It is time-consuming, unrelated to yield, and difficult to reflect the actual severity, especially when it comes to large-scale remote-sensing monitoring. We integrated the cotton VW progression mechanism exploring the potential of leaf VW severity in various leaf types at different layers of cotton to indicate yield loss. Based on all main stem leaves (MLs) in cotton leaf layers 1-3 and fruit branch leaves (FLs) in layers 1-5, we proposed a practical grading method for cotton VW associated with yield loss and suitable for remote sensing monitoring, termed the Eight-Position Grading method (EPG). The results indicated FL exhibited stronger correlation with yield compared to ML, and MLs in layers 1-3 and FLs in layers 1-5 effectively indicated yield loss due to VW. EPG was compared with the technical specifications for VW severity assessment in China (GB) and its associated grading methods, demonstrating performance with a 12 % yield loss gradient while correcting overestimation in GB grading. The remote sensing applicability of EPG was theoretically validated using PROSPECT_D and mSCOPE. Field remote sensing experiment confirmed that EPG achieved preferable accuracies in estimating VW severity (R2 = 0.76, RPD = 2.06) and demonstrated a strong correlation with yield (R2 = 0.53). This study offers a simple and practical method for scientifically assessing VW severity and estimating yield loss.

Orchard classification is vital for precision agriculture, enabling crop monitoring and sustainable land management. Despite progress in classification techniques, deep learning based spectral matching for orchard classification remains underexplored. This study introduces a deep learning based novel spectral matching method, the Time-Aware Siamese Network (TASN), to improve orchard classification through Spectrotemporal Signatures (STS) derived from multi-temporal vegetation indices (VI) stored in Multidimensional Data (MDD) format. Focusing on Khairpur, Pakistan, we fused Landsat-8/9 and Sentinel-2 datasets using Google Earth Engine (GEE) to create cloud-free monthly composites. Three VI were computed and different seasonal data cubes were generated to extract STS. Wavelet-based smoothing effectively reduces noise and enhances STS quality. TASN outperformed Spectral Angle Mapper (SAM), achieving 93% accuracy with UNVI based Winter-Spring-Summer (WSS) and Spring-Summer-Autumn (SSA) data cubes and 94% with Winter-Spring-Summer-Autumn (WSSA). Notably, UNVI using TASN slightly surpassed NDVI and EVI across all data cubes. The model's dual Long Short Term Memory (LSTM) layers effectively distinguished spectrally similar orchards, such as mango and banana, that challenge the SAM approach. These outcomes highlight the potential of deep learning architecture for spectral matching, providing a robust framework for improving orchard classification and advancing agricultural applications.

This paper presents a comprehensive analysis of the photometric system of the University of Chinese Academy of Sciences 70 cm Telescope located at the Yan-qi Lake campus of the University of Chinese Academy of Sciences.We evaluated the linearity,bias stability,and dark current of the camera.Utilizing the Johnson-Cousins Blue-Visible-Red-Infrared filter system and an Andor DZ936 charge-coupled device camera,we conducted extensive observations of Landolt standard stars to determine the color terms,atmospheric extinction coefficients,photometric zero-points,and the sky background brightness.The results indicate that this telescope demonstrates excellent performance in photometric calibration and good system performance overall,meeting the requirements for limited scientific research and teaching purposes.

This study evaluates the effectiveness of hyperspectral data to retrieve chlorophyll a (Chl-a) concentrations using various Machine Learning (ML) methods, specifically to determine whether spectral reflectance can provide accurate estimations of Chl-a. The study aims to address the gap in understanding how hyperspectral measurements correlate with Chl-a concentrations and to explore the potential for improving water quality assessment by accurately estimating Chl-a concentrations, which is essential for environmental monitoring, especially in aquatic ecosystems. The method proposed is evaluated using different Chl-a concentrations defined by the experiment design using Rhodamine B. The main reason for preparing pre-defined solutions of Chl-a is to verify the sensitivity of spectral measurements to Chl-a concentrations. In this paper, we aim to measure the pure signature of the Chl-a in which spectral reflectance of each Chl-a concentration is measured with 10 replicates by the spectrometer HS-1000WFL3. Six ML methods were investigated; (i) the multilayer perceptron artificial neural network (MLPNN), (ii) the support vector regression (SVR), (iii) the random forest regression (RFR), (iv) the Gaussian process regression (GPR), (v) Relevance Vector Machine (RVM) and (vi) Extreme Gradient Boosting (XGboost). 70