Hyperspectral sensing for turbid water quality monitoring in freshwater rivers: Empirical relationship between reflectance and turbidity and total solids

Estimating turbidity concentrations in highly dynamic rivers using Sentinel-2 imagery in Google Earth Engine: Case study of the Godavari River, India

Turbidity is an essential biogeochemical parameter for water quality management because it shapes the physical landscape and regulates ecological systems. It varies spatially and temporally across large water bodies, but monitoring based on point-source field observations remains a difficult task in developing countries due to the need for logistics and costs. In this study, we present a novel semi-analytical approach for estimating turbidity from remote sensing reflectance ( R rs ) in moderate to highly turbid waters in the lower part of the Godavari River (i.e., locations near Rajahmundry). The proposed method includes two sub-algorithms-Normalized Difference Turbidity Index (NDTI) and semi-empirical single-band turbidity ( T s ) algorithm-to retrieve spectral reflectance information corresponding to the study locations for turbidity modeling. Sentinel-2 Multi-Spectral Imager data have been used to quantify the turbidity in the Google Earth Engine (GEE) platform. The correlation analysis was observed between spectral reflectance values and in situ turbidity data using cubic polynomial regression equations. The results indicated that the T s , which uses the only red-edge wavelength, identified turbidity as the most accurate across all locations (highest R2 = 0.91, lowest RMSE = 0.003), followed by NDTI (highest R2 = 0.85, lowest RMSE = 0.05), respectively. The remote sensing data application provides a better way to monitor turbidity at large spatio-temporal scales in attaining the water quality standards of the Godavari River.

UAV and satellite remote sensing for inland water quality assessments: a literature review

High spatial and temporal resolution data is crucial to comprehend the dynamics of water quality fully, support informed decision-making, and allow efficient management and protection of water resources. Traditional in situ water quality measurement techniques are both time-consuming and labor-intensive, resulting in databases with limited spatial and temporal frequency. To address these challenges, satellite-driven water quality assessment has emerged as an efficient and effective solution, offering comprehensive data on larger-scale water bodies. Numerous studies have utilized multispectral and hyperspectral remote sensing data from various sensors to assess water quality, yielding promising results. However, the recent popularity of unmanned aerial vehicle (UAV) remote sensing can be attributed to its high spatial and temporal resolution, flexibility, ability to capture data at different times of day, and relatively low cost compared to traditional platforms. This study presents a comprehensive review of the current state of the art in monitoring water quality in small inland water bodies using satellite and UAV remote sensing data. It encompasses an overview of atmospheric correction algorithms and the assessment of different water quality parameters. Furthermore, the review addresses the challenges associated with monitoring water quality in these bodies of water and emphasizes the potential of UAVs to overcome these challenges by providing accurate and reliable data.

UAV-based remote sensing of turbidity in coastal environment for regulatory monitoring and assessment

The adoption of Unmanned Aerial Vehicle (UAV) remote sensing for the regulatory monitoring of turbidity plumes induced by land reclamation operations remains a difficult task. Compared to UAV remote sensing on ambient turbidity in estuaries and rivers, such monitoring of construction-induced turbidity plumes requires significantly higher spatial resolutions and accuracy as well as wider turbidity ranges with nonlinear reflectance. In this study, a pilot-scale deployment of UAV-based hyperspectral sensing is carried out for this objective, with specific new elements developed to overcome the challenges and minimise the uncertainties involved. In particular, Machine learning (ML) models for the turbidity determination were trained by the large dataset collected to better capture the non-linearity of the relationship between the water leaving reflectance and turbidity level. The models achieve a good accuracy with a R2 score of 0.75 that is deemed acceptable in view of the uncertainties associated with construction and land reclamation work.

An improved hyperspectral sensing approach for the rapid determination of copper ion concentrations in water environment using short-wavelength infrared spectroscopy

Copper ion is one of the hazardous pollutants often present in industrial wastewater or acid mine drainage that is regarded as a primary environmental challenge. Hyperspectral remote sensing has a long tradition in water quality monitoring. However, its application in heavy metal detection is relatively similar, and the detection is highly influenced by water turbidity or total suspended matter (TSM), requiring research efforts to improve accuracy and generalize the applicability of this technique. In this study, the use of simple filtration (pore size of 0.7 μm) for sample pretreatment to improve hyperspectral remote sensing of copper ion concentrations (Cu, 100-1000 mg/L) in water samples is proposed. A wide variety of water samples, including as-prepared and field (fish pond and river water) samples, were investigated to validate the developed method. Spectral data containing sensitive bands characterized in the range of 900-1100 nm were first preprocessed with logarithm transformation, followed by quantitative prediction model development using stepwise multivariate linear regression (SMLR) with the most sensitive wavebands at around 900 nm and 1080 nm. Satisfactory prediction performance for Cu ions was found for turbid water samples (TSM greater than approximately 200 mg/L) after simple filtration pretreatment, suggesting that pretreatment removed suspended solids in the mixtures and enhanced the spectral features of Cu ions in the model. Moreover, good agreement between the laboratory results and the field samples (adjusted R2 > 0.95 and NRMSE <0.15) highlights the suitability of the developed model and filtration pretreatment for obtaining relevant information for the rapid determination of Cu ion concentrations in complex water samples.

Assessment of water quality parameters in Muthupet estuary using hyperspectral PRISMA satellite and multispectral images

The continuous availability of spatial and temporal distributed data from satellite sensors provides more accurate and timely information regarding surface water quality parameters. Remote sensing data has the potential to serve as an alternative to traditional on-site measurements, which can be resource-intensive due to the time and labor involved. This present study aims in exploring the possibility and comparison of hyperspectral and multispectral imageries (PRISMA) for accurate prediction of surface water quality parameters. Muthupet estuary, situated on the south side of the Cauvery River delta on the Bay of Bengal, is selected as the study area. The remote sensing data is acquired from the PRISMA hyperspectral satellite and the Sentinel-2 multispectral instrument (MSI) satellite. The in situ sampling from the study area is performed, and the testing procedures are carried out for analyzing different water quality parameters. The correlations between the water sample results and the reflectance values of satellites are analyzed to generate appropriate algorithmic models. The study utilized data from both the PRISMA and Sentinel satellites to develop models for assessing water quality parameters such as total dissolved solids, chlorophyll, pH, and chlorides. The developed models demonstrated strong correlations with R² values above 0.80 in the validation phase. PRISMA-based models for pH and chlorophyll displayed higher accuracy levels than Sentinel-based models with R² > 0.90.

Use of Hyperspectral Reflectance and Water Quality Indices to Assess Groundwater Quality for Drinking in Arid Regions, Saudi Arabia

Combining hydrogeochemical characterization and a hyperspectral reflectance measurement can provide knowledge for groundwater security under different conditions. In this study, comprehensive examinations of 173 groundwater samples were carried out in Makkah Al-Mukarramah Province, Saudi Arabia. Physicochemical parameters, water quality indices (WQIs), and spectral reflectance indices (SRIs) were combined to investigate water quality and controlling factors using multivariate modeling techniques, such as partial least-square regression (PLSR) and principal component regression (PCR). To measure water quality status, the drinking water quality index (DWQI), total dissolved solids (TDS), heavy metal index (HPI), contamination degree (Cd), and pollution index (PI) were calculated. Standard analytical methods were used to assess nineteen physicochemical parameters. The typical values of ions and metals were as follows: Na2+ > Ca2+ > Mg2+ > K+, Cl− > SO42− > HCO3− > NO3− > CO32−; and Cu > Fe > Al > Zn > Mn > Ni, respectively. The hydrogeochemical characteristics of the examined groundwater samples revealed that Ca-HCO3, Na-Cl, mixed Ca-Mg-Cl-SO4, and Na-Ca-HCO3 were the main mechanisms governing groundwater chemistry and quality under the load of seawater intrusion, weathering, and water-rock interaction. According to the WQIs results, the DWQI values revealed that 2.5% of groundwater samples were categorized as excellent, 18.0% as good, 28.0% as poor, 21.5% as extremely poor, and 30.0% as unfit for drinking. The HPI and Cd values revealed that all groundwater samples had a low degree of contamination and better quality. Furthermore, the PI values showed that the groundwater resources were not affected by metals but were slightly affected by Mn in Wadi Fatimah due to rock–water interaction. Linear regression models demonstrated the significant relationships for the majority of SRIs paired with DWQI (R varied from −0.40 to 0. 75), and with TDS (R varied from 0.46 to 0.74) for the studied wadies. In general, the PLSR and PCR models provide better estimations for DWQI and TDS than the individual SRI. In conclusion, the grouping of WQIs, SRIs, PLSR, PCR, and GIS tools provides a clear image of groundwater suitability for drinking and its controlling elements.

Appraisal of Surface Water Quality of Nile River Using Water Quality Indices, Spectral Signature and Multivariate Modeling

Surface water quality management is an important facet of the effort to meet increasing demand for water. For that purpose, water quality must be monitored and assessed via the use of innovative techniques, such as water quality indices (WQIs), spectral reflectance indices (SRIs), and multivariate modeling. Throughout the Rosetta and Damietta branches of the Nile River, water samples were collected, and WQIs were assessed at 51 different distinct locations. The drinking water quality index (DWQI), metal index (MI), pollution index (PI), turbidity (Turb.) and total suspended solids (TSS) were assessed to estimate water quality status. Twenty-three physicochemical parameters were examined using standard analytical procedures. The average values of ions and metals exhibited the following sequences: Ca2+ > Na2+ > Mg2+ > K+, HCO32− > Cl− > SO42− > NO3− > CO3− and Al > Fe > Mn > Ba > Ni > Zn > Mo > Cr > Cr, respectively. Furthermore, under the stress of evaporation and the reverse ion exchange process, the main hydrochemical facies were Ca-HCO3 and mixed Ca-Mg-Cl-SO4. The DWQI values of the two Nile branches revealed that 53% of samples varied from excellent to good water, 43% of samples varied from poor to very poor water, and 4% of samples were unsuitable for drinking. In addition, the results showed that the new SRIs extracted from VIS and NIR region exhibited strong relationships with DWQI and MI and moderate to strong relationships with Turb. and TSS for each branch of the Nile River and their combination. The values of the R2 relationships between the new SRIs and WQIs varied from 0.65 to 0.82, 0.64 to 0.83, 0.41 to 0.60 and 0.35 to 0.79 for DWQI, MI, Turb. and TSS, respectively. The PLSR model produced a more accurate assessment of DWQI and MI based on values of R2 and slope than other indices. Furthermore, the partial least squares regression model (PLSR) generated accurate predictions for DWQI and MI of the Rosetta branch in the Val. datasets with an R2 of 0.82 and 0.79, respectively, and for DWQI and MI of the Damietta branch with an R2 of 0.93 and 0.78, respectively. Therefore, the combination of WQIs, SRIs, PLSR and GIS approaches are effective and give us a clear picture for assessing the suitability of surface water for drinking and its controlling factors.

The Integration of Multivariate Statistical Approaches, Hyperspectral Reflectance, and Data-Driven Modeling for Assessing the Quality and Suitability of Groundwater for Irrigation

Sustainable agriculture in arid regions necessitates that the quality of groundwater be carefully monitored; otherwise, low-quality irrigation water may cause soil degradation and negatively impact crop productivity. This study aimed to evaluate the quality of groundwater samples collected from the wells in the quaternary aquifer, which are located in the Western Desert (WD) and the Central Nile Delta (CND), by integrating a multivariate analysis, proximal remote sensing data, and data-driven modeling (adaptive neuro-fuzzy inference system (ANFIS) and support vector machine regression (SVMR)). Data on the physiochemical parameters were subjected to multivariate analysis to ease the interpretation of groundwater quality. Then, six irrigation water quality indices (IWQIs) were calculated, and the original spectral reflectance (OSR) of groundwater samples were collected in the 302–1148 nm range, with the optimal spectral wavelength intervals corresponding to each of the six IWQIs determined through correlation coefficients (r). Finally, the performance of both the ANFIS and SVMR models for evaluating the IWQIs was investigated based on effective spectral reflectance bands. From the multivariate analysis, it was concluded that the combination of factor analysis and principal component analysis was found to be advantageous to examining and interpreting the behavior of groundwater quality in both regions, as well as predicting the variables that may impact groundwater quality by illuminating the relationship between physiochemical parameters and the factors or components of both analyses. The analysis of the six IWQIs revealed that the majority of groundwater samples from the CND were highly suitable for irrigation purposes, whereas most of the groundwater from the WD can be used with some limitations to avoid salinity and alkalinity issues in the long term. The high r values between the six IWQIs and OSR were located at wavelength intervals of 302–318, 358–900, and 1074–1148 nm, and the peak value of r for these was relatively flat. Finally, the ANFIS and SVMR both obtained satisfactory degrees of model accuracy for evaluating the IWQIs, but the ANFIS model (R2 = 0.74–1.0) was superior to the SVMR (R2 = 0.01–0.88) in both the training and testing series. Finally, the multivariate analysis was able to easily interpret groundwater quality and ground-based remote sensing on the basis of spectral reflectance bands via the ANFIS model, which could be used as a fast and low-cost onsite tool to estimate the IWQIs of groundwater.

Combining Hydrogeochemical Characterization and a Hyperspectral Reflectance Tool for Assessing Quality and Suitability of Two Groundwater Resources for Irrigation in Egypt

Hyperspectral reflectance sensing provides a rapid and cost-effective technique for assessing the suitability of groundwater for irrigation by monitoring real-time changes in its quality at a large scale. In this study, we assessed the quality of 15 groundwater samples from El Fayoum depression in the Western Desert (WD) and 25 groundwater samples from the Central Nile Delta (CND) in Egypt using a traditional approach of the physiochemical parameters, irrigation water quality indicators (IWQIs), and hydrochemical facies. The spectral reflectance data of the water surface was used to build new simple reflectance indices (SRIs), and the performance of these indices for assessing IWQIs was compared with those by partial least square regression (PLSR) that was based on all SRIs or the full-spectrum ranges. Generally, the groundwater of the CND was fresher and more suitable for irrigation purposes than those of the WD. Based on the six IWQIs, ~6.7–60.0% and 85.0–100.0% of the groundwater samples of the WD and CND, respectively, were categorized as highly suitable for irrigation purposes. Based on hydrochemical facies, Na-Cl and Ca-HCO3 were dominant in the WD and CND, respectively, as well as the alkali earth metals (Na+ + K+), which significantly exceeded the alkaline earth metals (Ca2+ + Mg2+) in the WD, with the reverse for the CND. Most developed SRIs had a moderate, weak, and moderate to strong relationship with physiochemical parameters and IWQIs in the WD, CND, and across both regions, respectively. The PLSR models based on all SRIs provided a more accurate estimation of IWQIs in calibration and validation datasets than those based on full-spectrum ranges, and both PLSR models provided better estimation than the individual SRIs. These findings support the feasibility of using ground reflectance measurements as a fast and low-cost tool for the assessment and management of groundwater for irrigation in arid and semiarid regions.

Environmental Reservoirs of Vibrio cholerae: Challenges and Opportunities for Ocean-Color Remote Sensing

The World Health Organization has estimated the burden of the on-going pandemic of cholera at 1.3 to 4 million cases per year worldwide in 2016, and a doubling of case-fatality-rate to 1.8% in 2016 from 0.8% in 2015. The disease cholera is caused by the bacterium Vibrio cholerae that can be found in environmental reservoirs, living either in free planktonic form or in association with host organisms, non-living particulate matter or in the sediment, and participating in various biogeochemical cycles. An increasing number of epidemiological studies are using land- and water-based remote-sensing observations for monitoring, surveillance, or risk mapping of Vibrio pathogens and cholera outbreaks. Although the Vibrio pathogens cannot be sensed directly by satellite sensors, remotely-sensed data can be used to infer their presence. Here, we review the use of ocean-color remote-sensing data, in conjunction with information on the ecology of the pathogen, to map its distribution and forecast risk of disease occurrence. Finally, we assess how satellite-based information on cholera may help support the Sustainable Development Goals and targets on Health (Goal 3), Water Quality (Goal 6), Climate (Goal 13), and Life Below Water (Goal 14).

Comparison of satellite reflectance algorithms for estimating turbidity and cyanobacterial concentrations in productive freshwaters using hyperspectral aircraft imagery and dense coincident surface observations

We analyzed 37 satellite reflectance algorithms and 321 variants for five satellites for estimating turbidity in a freshwater inland lake in Ohio using coincident real hyperspectral aircraft imagery converted to relative reflectance and dense coincident surface observations. This study is part of an effort to develop simple proxies for turbidity and algal blooms and to evaluate their performance and portability between satellite imagers for regional operational turbidity and algal bloom monitoring. Turbidity algorithms were then applied to synthetic satellite images and compared to in situ measurements of turbidity, chlorophyll-a (Chl-a), total suspended solids (TSS) and phycocyanin as an indicator of cyanobacterial/blue green algal (BGA) abundance. Several turbidity algorithms worked well with real Compact Airborne Spectrographic Imager (CASI) and synthetic WorldView-2, Sentinel-2 and Sentinel-3/MERIS/OLCI imagery. A simple red band algorithm for MODIS imagery and a new fluorescence line height algorithm for Landsat-8 imagery had limited performance with regard to turbidity estimation. Blue-Green Algae/Phycocyanin (BGA/PC) and Chl-a algorithms were the most widely applicable algorithms for turbidity estimation because strong co-variance of turbidity, TSS, Chl-a, and BGA made them mutual proxies in this experiment.

Evaluation of Regression Analysis and Neural Networks to Predict Total Suspended Solids in Water Bodies from Unmanned Aerial Vehicle Images

The concentration of suspended solids in water is one of the quality parameters that can be recovered using remote sensing data. This paper investigates the data obtained using a sensor coupled to an unmanned aerial vehicle (UAV) in order to estimate the concentration of suspended solids in a lake in southern Brazil based on the relation of spectral images and limnological data. The water samples underwent laboratory analysis to determine the concentration of total suspended solids (TSS). The images obtained using the UAV were orthorectified and georeferenced so that the values referring to the near, green, and blue infrared channels were collected at each sampling point to relate with the laboratory data. The prediction of the TSS concentration was performed using regression analysis and artificial neural networks. The obtained results were important for two main reasons. First, although regression methods have been used in remote sensing applications, they may not be adequate to capture the linear and/or non-linear relationships of interest. Second, results show that the integration of UAV in the mapping of water bodies together with the application of neural networks in the data analysis is a promising approach to predict TSS as well as their temporal and spatial variations.

Simple and efficient dual-wavelength spectroscopy for the determination of organic matter in sewage sludge from wastewater treatment

In this study, a dual-wavelength spectroscopic method for rapid determination of organic matter in sludge was developed. The contents of the organic matter were calculated by determining the consumption of potassium dichromate (K₂Cr₂O₇) based on the production of trivalent chromium ions (Cr³⁺). Cr³⁺ could be determined by subtracting the absorption at 800 nm (spectral interference) from the absorption at 650 nm (only contributed by Cr³⁺). The results showed that the relative standard deviation in the test was less than 5%. The same set of samples was used and when the content of organic matter was more than 150 g kg⁻¹, the relative difference between the spectroscopic method and titration method was within 1%. Furthermore, the method does not require calibration based on the standard samples. In conclusion, the present method is simple, reliable, accurate and suitable for application in mass testing for sludge samples.

In this study, a dual-wavelength spectroscopic method for rapid determination of organic matter in sludge was developed.

The Assessment of Landsat-8 OLI Atmospheric Correction Algorithms for Inland Waters

The OLI (Operational Land Imager) sensor on Landsat-8 has the potential to meet the requirements of remote sensing of water color. However, the optical properties of inland waters are more complex than those of oceanic waters, and inland atmospheric correction presents additional challenges. We examined the performance of atmospheric correction (AC) methods for remote sensing over three highly turbid or hypereutrophic inland waters in China: Lake Hongze, Lake Chaohu, and Lake Taihu. Four water-AC algorithms (SWIR (Short Wave Infrared), EXP (Exponential Extrapolation), DSF (Dark Spectrum Fitting), and MUMM (Management Unit Mathematics Models)) and three land-AC algorithms (FLAASH (Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes), 6SV (a version of Second Simulation of the Satellite Signal in the Solar Spectrum), and QUAC (Quick Atmospheric Correction)) were assessed using Landsat-8 OLI data and concurrent in situ data. The results showed that the EXP (and DSF) together with 6SV algorithms provided the best estimates of the remote sensing reflectance (Rrs) and band ratios in water-AC algorithms and land-AC algorithms, respectively. AC algorithms showed a discriminating accuracy for different water types (turbid waters, in-water algae waters, and floating bloom waters). For turbid waters, EXP gave the best Rrs in visible bands. For the in-water algae and floating bloom waters, however, all water-algorithms failed due to an inappropriate aerosol model and non-zero reflectance at 1609 nm. The results of the study show the improvements that can be achieved considering SWIR bands and using band ratios, and the need for further development of AC algorithms for complex aquatic and atmospheric conditions, typical of inland waters.

Validating chlorophyll-a concentrations in the Lagos Lagoon using remote sensing extraction and laboratory fluorometric methods

Remote sensing data is a viable alternative for mapping pigment concentrations in water body, and consequently, the trophic. Chlorophyll-a (Chl-a) is present in all phytoplankton species. This study therefore uses laboratory fluorometric and remote sensing extraction methods for assessing chlorophyll-a concentration in the Lagos Lagoon. The fluorometer was calibrated with a commercially available chlorophyll-a standard before used in the laboratory to estimates chlorophyll-a concentration. Landsat 7 (ETM+) and Landsat 8 (OLI) were acquired for the remote sensing method. The Landsat data were first geometrically rectified. Then brightness values were converted to reflectance through the radiometric correction process. For the regression models, logarithmically transformed chlorophyll-a was used as the dependent variable. Single bands, band ratios and logarithmically transformed band ratios were the independent variables. R2 values were computed and evaluated. •Chlorophyll-a contributes to productive water bodies•laboratory fluorometric and remote sensing extraction methods•Landsat data acquired for the remote sensing method.

Quantifying suspended solids in small rivers using satellite data

The management of suspended solids and associated contaminants in rivers requires knowledge of sediment sources. In-situ sampling can only describe the integrated impact of the upstream sources. Empirical models that use surface reflectance from satellite images to estimate total suspended solid (TSS) concentrations can be used to supplement measurements and provide spatially continuous maps. However, there are few examples, especially in narrow, shallow and hydrologically dynamic rivers found in mountainous areas. A case study of the Didipio catchment in Philippines was used to address these issues. Four 5-m resolution RapidEye images, from between the years 2014 and 2016, and near-simultaneous ground measurements of TSS concentrations were used to develop a power law model that approximates the relationship between TSS and reflectance for each of four spectral bands. A second dataset using two 2-m resolution Pleiades-1A and a third using a 6-m resolution SPOT-6 image along with ground-based measurements, were consistent with the model when using the red band data. Using that model, encompassing data from all three datasets, gave an R² value of 65% and a root mean square error of 519mgL^-1. A linear relationship between reflectance and TSS exists from 1mgL^-1 to approximately 500mgL^-1. In contrast, for TSS measurements between 500mgL^-1 and 3580mgL^-1 reflectance increases at a generally lower and more variable rate. The results were not sensitive to changing the pixel location within the vicinity of the ground sampling location. The model was used to generate a continuous map of TSS concentration within the catchment. Further ground-based measurements including TSS concentrations that are higher than 3580mgL^-1 would allow the model to be developed and applied more confidently over the full relevant range of TSS.

Proposal of a Method to Determine the Correlation between Total Suspended Solids and Dissolved Organic Matter in Water Bodies from Spectral Imaging and Artificial Neural Networks

Water quality monitoring through remote sensing with UAVs is best conducted using multispectral sensors; however, these sensors are expensive. We aimed to predict multispectral bands from a low-cost sensor (R, G, B bands) using artificial neural networks (ANN). We studied a lake located on the campus of Unisinos University, Brazil, using a low-cost sensor mounted on a UAV. Simultaneously, we collected water samples during the UAV flight to determine total suspended solids (TSS) and dissolved organic matter (DOM). We correlated the three bands predicted with TSS and DOM. The results show that the ANN validation process predicted the three bands of the multispectral sensor using the three bands of the low-cost sensor with a low average error of 19%. The correlations with TSS and DOM resulted in R² values of greater than 0.60, consistent with literature values.

Data Field Modeling and Spectral-Spatial Feature Fusion for Hyperspectral Data Classification

Classification is a significant subject in hyperspectral remote sensing image processing. This study proposes a spectral-spatial feature fusion algorithm for the classification of hyperspectral images (HSI). Unlike existing spectral-spatial classification methods, the influences and interactions of the surroundings on each measured pixel were taken into consideration in this paper. Data field theory was employed as the mathematical realization of the field theory concept in physics, and both the spectral and spatial domains of HSI were considered as data fields. Therefore, the inherent dependency of interacting pixels was modeled. Using data field modeling, spatial and spectral features were transformed into a unified radiation form and further fused into a new feature by using a linear model. In contrast to the current spectral-spatial classification methods, which usually simply stack spectral and spatial features together, the proposed method builds the inner connection between the spectral and spatial features, and explores the hidden information that contributed to classification. Therefore, new information is included for classification. The final classification result was obtained using a random forest (RF) classifier. The proposed method was tested with the University of Pavia and Indian Pines, two well-known standard hyperspectral datasets. The experimental results demonstrate that the proposed method has higher classification accuracies than those obtained by the traditional approaches.