Split-window Approach Research Articles

LAND SURFACE TEMPERATURE INVESTIGATION AROUND IKOGOSI WARM SPRING NIGERIA USING LANDSAT8 DATA

Ikogosi warm spring (IWS) is among the most visited geothermal resource by tourists in Nigeria. On that basis, it has attracted so much attention from researchers using various geophysical methods, except the retrieval of the land surface temperature (LST) from remote sensing data. This work aimed at computing LST to delineate hot zone around Ikogosi geothermal resources. The split-window approach was used to compute the LST from Landsat 8 data. The interpretation of Landsat8 data revealed that the central region of the study area is of high LST, and the temperature then drops towards the southwest direction. The result also shows that the warm spring is situated around a region with high land surface temperature (about 29 °C) which is an indication of a geothermal reservoir. The supervised classification of the LST yields two zones of the high density of pixels with high temperature, hot spot zones. The hot spot zone west of IWS is believed to be the heat source of IWS as it has high LST, and it is closer to IWS while the hot spot zone NW of IWS shows an indication of a viable geothermal resource, high LST

Automated water surface temperature retrieval from Landsat 8/TIRS

Satellite remote sensing of Land and Water Surface Temperature (L/WST) has many applications in studies of terrestrial and aquatic ecology. Retrieval of L/WST requires a well calibrated radiometer and an accurate atmospheric correction. In the present study, the performance of the Thermal InfraRed Sensor (TIRS) on board Landsat 8 is evaluated for the retrieval of L/WST. libRadtran is used to retrieve atmospheric correction parameters based on atmospheric profiles of relative humidity and temperature from three global atmospheric models. Performance of single band retrievals is compared to typical MODTRAN results from the Atmospheric Correction Parameter Calculator (ACPC) and a split-window approach. A multi-temporal land masking method using imagery from the Operational Land Imager (OLI) on board Landsat 8 is demonstrated, and is used to automatically classify imagery in the matchup dataset in three classes of cloud cover. Two sources of in situ data covering the Belgian Coastal Zone (BCZ) are used for validation of the L/WST product: (1) fixed locations in the Flemish Banks measurement network and (2) underway data from regular RV Belgica campaigns. In the present study the single band methods outperformed the split-window approach, and consistent retrievals are found for the MODTRAN and libRadtran simulations. Typical single band surface temperature retrievals in quasi cloud-free conditions have Root Mean Squared Differences (RMSD) of 0.7 K and 1 K for Bands 10 and 11 with low bias, depending on the method and atmospheric profile source. For imagery with scattered clouds, RMSD values increase to 1 K and 2 K respectively with an approximately 0.5 K cold bias, likely caused by cloud proximity. The calibration efforts combined into Collection 1 allows for accurate absolute surface temperature retrievals from B10 on Landsat 8/TIRS for homogeneous targets with known emissivity, such as liquid water. The method is adapted to global processing and can be used for Land Surface Temperature retrieval with a suitable source of emissivity data.

Spatial and multi-depth temporal soil temperature assessment by assimilating satellite imagery, artificial intelligence and regression based models in arid area

Soil temperature is a noteworthy factor, particularly in farming and land treatment of organic wastes, since development of organic system is firmly controlled by soil temperature. Furthermore, soil temperature has impacts on the physical, substance, and microbiological forms that occur in soil. In this study, for estimation of multi-depth soil temperature, the most reliable combination of input variables for model development was achieved using gamma test technique. The most suitable training and testing data length i.e. training data set 11,529 (01 January 1975 to 25 July 2006; 81%) and testing data set 2716 (26 July 2006 to 31 December 2013; 19%) of the most efficient input variables were determined by means of M-test technique. Multi-depth soil temperature model was achieved by means of multiple linear regression (MLR), multilayer perceptron (MLP) and adaptive neuro-fuzzy inference system (ANFIS) techniques. After comparing results of all models, it was found that the ANFIS technique with generalized bell membership function had the best performing results when compared to MLR and MLP for all cases i.e. 5 cm, 10 cm and 20 cm depths of soil. Potential of gamma test technique was evaluated by means of sensitivity analysis. The spatial variation of land surface temperature (LST) was assessed by remote sensing (RS) and GIS technique and also the qualitative and quantitative performances were evaluated for determining the reliable and accurate spatial variation of LST. In qualitative evaluation, it was observed that the split-window (SW) approach overestimated the LST with coefficient of determination 0.8654. According to the quantitative evaluation based on root mean square error, correlation coefficient and coefficient of efficiency values between observed and estimated LST, the spatial variation of LST was found to be accurate over the Udaipur City, Rajasthan.

Land surface temperature estimation using split window approach over US Nagar district of Uttarakhand state, India

To estimate land surface temperature (LST) has an important role for agriculture as well as global change of climate, growth of vegetation and glacier melting. It combines the results of all surface atmosphere interactions and energy fluxes between the surface and the atmosphere. Now-a-days, estimation of temperature of land surface is being calculated with the help of satellite images containing thermal infrared band. Though land surface temperature derived from satellite, could be a beneficial complement to conventional land surface temperature data sources. This research, proposed a methodology for determining land surface temperature through using a structured mathematical algorithm viz., split window (SW) algorithm. Split window algorithm has been used on LANDSAT 8 with operational land imager i.e. OLI sensor and thermal infrared sensor i.e. TIRS dataset of Udham Singh Nagar district. TIRS shows two thermal bands i.e. band 10 and band 11. SW approach requires brightness temperature value of both band 10 and band 11 as well as land surface emissivity which is calculated from OLI bands i.e. NIR and Red, for the estimation of land surface temperature. The spectral radiance was determined using thermal infrared bands i.e. band 10 and band 11. Emissivity was calculated by using normalized difference vegetation index i.e. NDVI threshold technique for which OLI bands 2, 3, 4 and 5 were utilized. SW approach uses brightness temperature of two bands of thermal infrared, mean and difference in land surface emissivity for estimating land surface temperature. In this paper, 6 Dec. 2015 date was selected as an example to show the approach of using SW technique to estimate the LST of Udham Singh Nagar district of Uttarakhand state in India.

Validation of the modified Becker’s split-window approach for retrieving land surface temperature from AVHRR

To further verify the modified Becker’s split-window approach for retrieving land surface temperature (LST) from long-term Advanced Very High Resolution Radiometer (AVHRR) data, a cross-validation and a radiance-based (R-based) validation are performed and examined in this paper. In the cross-validation, 3481 LST data pairs are extracted from the AVHRR LST product retrieved with the modified Becker’s approach and compared with the Moderate Resolution Imaging Spectroradiometer (MODIS) LST product (MYD11A1) for the period 2002–2008, relative to the positions of 548 weather stations in China. The results show that in most cases, the AVHRR LST values are higher than the MYD11A1. When the AVHRR LSTs are adjusted with a linear regression, the values are close to the MYD11A1, showing a good linear relationship between the two datasets (R 2 = 0.91). In the R-based validation, comparison is made between AVHRR LST retrieved from the modified Becker’s approach and the inversed LST from the Moderate Resolution Transmittance Model (MODTRAN) consolidated with observed temperature and humidity profiles at four radiosonde stations. The results show that the retrieved AVHRR LST deviates from the MODTRAN inversed LST by–1.3 (–2.5) K when the total water vapor amount is less (larger) than 20 mm. This provides useful hints for further improvement of the LST retrieval algorithms’ accuracy and consistency.

Investigating the use of the Saharan dust index as a tool for the detection of volcanic ash in SEVIRI imagery

Despite the similar spectral signatures of ash and desert dust, relatively little has been done to explore the application of dust detection techniques to the problem of volcanic ash detection. The Saharan dust index (SDI) is routinely implemented for dust monitoring at some centres and could be utilised for volcanic ash detection with little computational expense, thereby providing a product that forecasters already have some familiarity with to complement the suite of existing ash detection tools. We illustrate one way in which the index could be implemented for the purpose of ash detection by applying it to three scenes containing volcanic ash from the 2010 Eyjafjallajökull eruption, Iceland and the 2011 eruption of Puyehue, Chile. It was also applied to an image acquired over Etna in January 2011, where a volcanic plume is clearly visible but is unlikely to contain any ash. These examples demonstrate the potential of the SDI as a tool for ash monitoring under different environmental and atmospheric conditions. In addition to presenting a valuable qualitative product to aid monitoring, this work includes a quantitative assessment of the detection skill using a manually constructed expert ash mask. The optimum implementation of any technique is likely to be dependent on both atmospheric conditions and on the properties of the imaged ash (which is often unknown in a real-time situation). Here we take advantage of access to a ‘truth’ rarely available in a real-time situation and calculate an ash mask based on the optimum threshold for the specific scene, which is then used to demonstrate the potential of the SDI. The SDI mask is compared to masks calculated from a simplistic implementation of the more traditional split window method, again exploiting our access to the ‘truth’ to set the most appropriate threshold for each scene, and to a probabilistic method that is implemented without reference to the ‘truth’ and which provides useful insights into the likely cloud-/ash-contamination of each pixel. Since the sensitivity of the SDI and split window methods to the tailored thresholds was not tested (such tailoring is unlikely to be possible in a real situation), this study presents the maximum anticipated skill for the SDI in the context of the maximum skill anticipated for the split window method, although both are likely to be lower in a real-time situation. The results for the SDI are comparable to those of the other methods, with a true skill score of 80.02% for the Eyjafjallajökull night-time scene (compared to 88.81% and 46.63% for the split window and probabilistic method respectively) and 90.06% for the Eyjafjallajökull day-time scene (compared to 97.61% and 56.96%). For the Puyehue image, the SDI resulted in a true skill score of 74.85%, while the split window approach achieved 99.62%. These results imply that the SDI, which is already implemented operationally at some centres for dust detection, could be a useful complement to existing ash monitoring techniques.

A modified Becker’s split-window approach for retrieving land surface temperature from AVHRR and VIRR

In order to provide a long time-series, high spatial resolution, and high accuracy dataset of land surface temperature (LST) for climatic change research, a modified Becker and Li’s split-window approach is proposed in this paper to retrieve LST from the measurements of Advanced Very High Resolution Radiometer (AVHRR) onboard National Oceanic and Atmospheric Administration (NOAA)-7 to -18 and the Visible and InfraRed Radiometer (VIRR) onboard FY-3A. For this purpose, the Moderate Resolution Transmittance Model (MODTRAN) 4.1 was first employed to compute the spectral radiance at the top of atmosphere (TOA) under a variety of surface and atmosphere conditions. Then, a temperature dataset consists of boundary temperature Ts (which is one of the input parameters to MODTRAN), and channels 4 and 5 brightness temperatures (T4 and T5) were constructed. Note that channels 4 and 5 brightness temperatures were simulated from the MODTRAN output spectral radiance by convolving them with the spectral response functions (SRFs) of channels 4 and 5 of AVHRRs and VIRR. The coefficients of modified Becker and Li’s split-window approach for various AVHRRs and VIRR were subsequently regressed based on this temperature dataset using the least square method. As an example of validation, one AVHRR satellite image over Beijing acquired at 0312 UTC 27 April 2008 by AVHRR onboard NOAA-17 was selected to retrieve the LST image using the modified Becker and Li’s approach. The comparison between this LST image and that from the MODIS level-2 LST product provided by the University of Tokyo in Japan indicates that the correlation coefficient is 0.88, the bias is 0.6 K, and the root mean square deviation (RMSD) is 2.1 K. Furthermore, about 70% and 37% pixels in the LST difference image, which is the result of retrieved LST image from AVHRR minus the corresponding MODIS LST image, have the values within ±2 and ±1 K, respectively.

Gazing at Cirrus Clouds for 25 Years through a Split Window. Part I: Methodology

Abstract This paper demonstrates that the split-window approach for estimating cloud properties can improve upon the methods commonly used for generating cloud temperature and emissivity climatologies from satellite imagers. Because the split-window method provides cloud properties that are consistent for day and night, it is ideally suited for the generation of a cloud climatology from the Advanced Very High Resolution Radiometer (AVHRR), which provides sampling roughly four times per day. While the split-window approach is applicable to all clouds, this paper focuses on its application to cirrus (high semitransparent ice clouds), where this approach is most powerful. An optimal estimation framework is used to extract estimates of cloud temperature, cloud emissivity, and cloud microphysics from the AVHRR split-window observations. The performance of the split-window approach is illustrated through the diagnostic quantities generated by the optimal estimation approach. An objective assessment of the performance of the algorithm cloud products from the recently launched space lidar [Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation/Cloud-Aerosol Lidar with Orthogonal Polarization (CALIPSO/CALIOP)] is used to characterize the performance of the AVHRR results and also to provide the constraints needed for the optimal estimation approach.

Analysis of the NPOESS VIIRS land surface temperature algorithm using MODIS data

The Visible Infrared Imaging Radiometer Suite (VIIRS) will replace the National Oceanic and Atmospheric Administration's (NOAA) Advanced Very High Resolution Radiometer (AVHRR) and the National Aeronautics and Space Administration's (NASA) Moderate Resolution Imaging Spectrometer (MODIS) as the nation's wide-swath multispectral sensor following the launch of the National Polar Orbiting Environmental Sensor Suite (NPOESS) Preparatory Project (NPP) in about 2008. Through the follow-on NPOESS program, VIIRS data will be the primary source of systematic land remote sensing products until about 2022. Together, the AVHRR/MODIS/VIIRS satellite series will provide a critical long-term data record for Earth studies. In the NPP/NPOESS program, the product algorithms and production environments are developed and managed by private industry. This is a significant change from the current NOAA and NASA programs, and NPOESS algorithm deviations from successful heritage approaches (e.g., Earth Observation System) warrant comprehensive independent testing. The current baseline VIIRS land surface temperature (LST) algorithm represents one such deviation. In the present study, we evaluated the VIIRS LST by adapting it for use with 60 scenes of MODIS Level 1b radiance data. Algorithm coefficients were derived from MODTRAN4 radiative transfer model simulations. Using the validated MODIS LST (MYD11/spl I.bar/L2) product as a reference, we found that precision errors in the VIIRS dual split window (DSW) algorithm (the current "main" approach) significantly exceed those of the VIIRS split window (SW) algorithm (the current "backup" approach) in both daytime and nighttime conditions. Performance of both is better for nighttime cases than for daytime cases with all surface types. We attribute the larger errors in the DSW approach to its use of short middle infrared wavelengths which, compared to thermal infrared wavelengths, exhibit greater variability in surface emissivity and susceptibility to solar contamination. We conclude that a traditional SW algorithm, such as the current VIIRS backup algorithm, would provide superior performance to the DSW approach. An SW approach would also provide more seamless continuity with heritage products. Although this is not an NPOESS requirement, it is a key objective for multimission climate data records and Earth system data records.

In-flight validation and recovery of water surface temperature with Landsat-5 thermal infrared data using an automated high-altitude lake validation site at Lake Tahoe

The absolute radiometric accuracy of the thermal infrared band (B6) of the Thematic Mapper (TM) instrument on the Landsat-5 (L5) satellite was assessed over a period of approximately four years using data from the Lake Tahoe automated validation site (California-Nevada). The Lake Tahoe site was established in July 1999, and measurements of the skin and bulk temperature have been made approximately every 2 min from four permanently moored buoys since mid-1999. Assessment involved using a radiative transfer model to propagate surface skin temperature measurements made at the time of the L5 overpass to predict the at-sensor radiance. The predicted radiance was then convolved with the L5B6 system response function to obtain the predicted L5B6 radiance, which was then compared with the radiance measured by L5B6. Twenty-four cloud-free scenes acquired between 1999 and 2003 were used in the analysis with scene temperatures ranging between 4/spl deg/C and 22/spl deg/C. The results indicate L5B6 had a radiance bias of 2.5% (1.6/spl deg/C) in late 1999, which gradually decreased to 0.8% (0.5/spl deg/C) in mid-2002. Since that time, the bias has remained positive (predicted minus measured) and between 0.3% (0.2/spl deg/C) and 1.4% (0.9/spl deg/C). The cause for the cold bias (L5 radiances are lower than expected) is unresolved, but likely related to changes in instrument temperature associated with changes in instrument usage. The in situ data were then used to develop algorithms to recover the skin and bulk temperature of the water by regressing the L5B6 radiance and the National Center for Environmental Prediction (NCEP) total column water data to either the skin or bulk temperature. Use of the NCEP data provides an alternative approach to the split-window approach used with instruments that have two thermal infrared bands. The results indicate the surface skin and bulk temperature can be recovered with a standard error of 0.6/spl deg/C. This error is larger than errors obtained with other instruments due, in part, to the calibration bias. L5 provides the only long-duration high spatial resolution thermal infrared measurements of the land surface. If these data are to be used effectively in studies designed to monitor change, it is essential to continue to monitor instrument performance in-flight and develop quantitative algorithms for recovering surface temperature.

Estimation of land surface temperature from a Geostationary Operational Environmental Satellite (GOES‐8)

Two algorithms are developed and applied to observations from the Geostationary Operational Environmental Satellite (GOES) to enable frequent estimate of Land Surface Temperature (LST) representing the diurnal cycle. The derived LSTs are evaluated against a wide range of ground observations. Both algorithms are based on radiative transfer theory; one is similar to the classical split window approach used for deriving Sea Surface Temperature (SST), while the other is a three‐channel algorithm. The three‐channel LST algorithm aims to improve atmospheric correction by utilizing the characteristics of the middle‐infrared (MIR) band. Effects of both the atmosphere and the surface emissivity are accounted for. The simulations from the proposed algorithms are compared with previously developed generalized split window algorithm, and a split window algorithm with water vapor correction. During daytime, the proposed new split window algorithm gives the best LST retrievals, while during nighttime, the proposed three‐channel algorithm gives the best retrievals, both within a Root Mean Square (RMS) error of less than 1 K and without a significant bias. Evaluations against the Atmospheric Radiation Measurement (ARM) observations of radiometric surface temperatures and Surface Radiation Network (SURFRAD) observations of outgoing long wave (LW) radiation indicate that LST can be determined from the actual GOES‐8 observations within an RMS accuracy of about 1–2 K, standard error of about 1 K, and bias of less than 1 K. When evaluated against the North Carolina Agricultural Research Service (NCARS) soil temperature as observed at depth of 8 in. and against air temperature observations, the amplitude of the retrieved LST is found to be significantly greater than that of the observed soil temperature, lower than the nighttime air temperature, and higher than the daytime air temperature. When the soil observations are “corrected” to account for the depth difference, they are in good agreement with the LST retrieved from the satellite observations. This indicates that observations of soil temperature, which are more readily available than measurements of “skin” temperatures, can be useful in evaluating satellite‐based estimates. The LST retrieved from both of the proposed algorithms and from a NOAA/NESDIS algorithm, are generally very close to the converted skin temperature from the SURFRAD surface outgoing LW radiation. In most cases, the newly proposed algorithm shows better agreement with ground observations.

Thermal remote sensing of near-surface water vapor

In this study, four approaches to estimate atmospheric water vapor from Advanced Very High Resolution Radiometer (AVHRR) observations were tested with data from the Boreal Ecosystem–Atmosphere Study (BOREAS) and the Oklahoma Mesonetwork. The approaches studied were (i) the split-window difference of the thermal channels (Channel 4: 10.3–11.3 μm and Channel 5: 11.5–12.5 μm) by Dalu [Int. J. Remote Sens. 7 (1986) 1089.], (ii) the ratio of variances by Jedlovec [J. Appl. Meteorol. 29 (1990) 863.], (iii) the regression slope by Goward et al. [Ecol. Appl. 4 (1994) 322.], and (iv) a look-up table derived from radiative transfer model output. Although these techniques were primarily developed to estimate total column precipitable water, we used them to estimate near-surface water vapor, within a few meters of the surface. Near-surface water vapor is needed for hydrologic and biospheric modeling. Analysis showed the total column precipitable water to be highly correlated ( r 2=.79) with near-surface absolute humidity for clear-sky conditions at the BOREAS and the Oklahoma study sites. Correlation of all the retrieval techniques with ground observations was very low. For the split-window approach, water vapor can only be estimated on a per pixel basis and is ambiguous for anything but a single site. The regression slope and variance ratio techniques showed very little correlation with ground observations with r 2=.02 when compared with data from BOREAS, and .17 for the variance ratio and .24 for the regression slope when compared with Mesonet data. The spatial variability of water vapor across the landscape hampers the use of these contextual approaches. The highest correlation was for the look-up table approach, with r 2=.36 when compared with data from the BOREAS site. The look-up table was applied using AVHRR Channels 4 and 5 brightness temperatures, surface temperature, and near-surface air temperature. Surface temperature and air temperature were both estimated from the satellite readings. Combining the satellite data with air temperature measured at meteorological ground stations improved the correlation to .50. The relatively low r 2 values were at least partly due to spatial and temporal mismatches between surface and satellite measurements. Simulation of Moderate Resolution Imaging Spectrometer (MODIS) thermal Channels 29 (8.4–8.7 μm), 31 (10.78–11.28 μm), and 32 (11.77–12.27 μm) brightness temperatures showed that Channels 31 and 32 provide similar information as AVHRR Channels 4 and 5. The additional thermal information provided by Channel 29 shows promise for future water vapor detection efforts.

Impact of AVHRR filter functions on surface temperature estimation from the split window approach

With the development of long-term remotely-sensed data sets through the National Oceanic and Atmospheric Administration (NOAA) and National Aeronautics and Space Administration (NASA) Pathfinder programme, decadal length studies of the land surface are being conducted. However, data from five different Advanced Very High Resolution Radiometer (AVHRR) instruments and thus five different filter functions for each thermal channel must be used. If variations in sensor filter functions are not taken into account when the split window technique is applied, considerable error in the estimation of land surface temperature can be produced. For example, the Becker and Li (1990) equation was designed to be used with data from the AVHRR instrument on the NOAA-9 satellite. If it is applied to data from an AVHRR sensor other than NOAA-9's, the error in surface temperature can be as large as 2.3 degK.

Atmospheric correction of NOAA-11 NDVI data in the arid rangelands of Central Australia

Two empirical approaches to atmospheric correction of NOAA-11 NDVI data are illustrated for the arid lands of Central Australia, using (1) radiosonde data and (2) temperature differences ( T4-T5 ) between bands 4 and 5 of NOAA (split-window). Scattering angle is also shown to influence NDVI data in the Central Australian region. The split-window approach provides the better relationship and can be summarized by a linear formula. The use of preflight or sensor degradation calibrations of bands 1 and 2 have no significant influence on the strength of atmospheric correction procedures tested.

Atmospheric correction and determination of sea surface temperature in midlatitudes from NOAA-AVHRR data

In this paper we derive and validate a split-window algorithm for retrieving the sea surface temperature ( SST) in midlatitude atmospheric conditions. First we have analyzed thephysical fundamentals of the split-window approach. Then an algorithm has been theoretically derived by means of statistical regression over satellite measurement simulations, and it has been tuned for real NOAA-AVHRR measurements using in situ and coincident satellite data. The accuracy achieved for SST is 0.5 K, which is the limit accuracy that can be obtained from AVHRR measurements over midlatitudes. On the other side, we have proposed a strategy for applying the split-window algorithm without significantly increasing the noise level. This has the advantage of enhancing real thermal gradients and removing ficticious gradients introduced by the atmosphere.