Get Our e-AlertsSubmit Manuscript
Journal of Remote Sensing / 2022 / Article

Research Article | Open Access

Volume 2022 |Article ID 9845432 | https://doi.org/10.34133/2022/9845432

Shanshan Du, Xinjie Liu, Jidai Chen, Liangyun Liu, "Prospects for Solar-Induced Chlorophyll Fluorescence Remote Sensing from the SIFIS Payload Onboard the TECIS-1 Satellite", Journal of Remote Sensing, vol. 2022, Article ID 9845432, 9 pages, 2022. https://doi.org/10.34133/2022/9845432

Prospects for Solar-Induced Chlorophyll Fluorescence Remote Sensing from the SIFIS Payload Onboard the TECIS-1 Satellite

Received23 Jun 2022
Accepted08 Sep 2022
Published23 Sep 2022

Abstract

The importance of solar-induced chlorophyll fluorescence (SIF) to monitoring vegetation photosynthesis has attracted much attention from the ecological and remote sensing research communities. Space-borne SIF products have been obtained owing to the rapid development of atmospheric satellites in recent years. The SIF Imaging Spectrometer (SIFIS) is a payload onboard the upcoming Terrestrial Ecosystem Carbon Inventory Satellite (TECIS-1) that is specifically designed for SIF monitoring. We conducted an in situ experiment to evaluate the performance of SIFIS on spectral measurement and SIF retrieval through comparison to the commercial spectrometer QE Pro. Disregarding the spatiotemporal mismatch between the collected measurements of the two spectrometers, the radiance spectra obtained synchronously by SIFIS and QE Pro showed a high level of consistency. The SIF retrieval, normalized difference vegetation index (NDVI), and near-infrared radiance of vegetation (NIRvR) results for a push-broom image shows consistent spatial distributions over both vegetated and nonvegetated surfaces. A quantitative comparison was conducted by strictly filtering matching pixels. For the far-red band, a high correlation was obtained between the SIF retrieval performances of SIFIS and QE Pro with and . However, a relatively poor correlation was observed for the red band with an value of 0.23 and an RMSE of 0.26 mWm−2sr-−1nm−1. Despite the large uncertainties associated with this experiment, the results indicate that TECIS-1 should offer a reliable SIF monitoring performance after its launch.

1. Introduction

Solar-induced chlorophyll fluorescence (SIF) originates from chlorophyll molecules in a plant that absorb photosynthetically active radiation, and it is tightly linked with vegetation photosynthesis [13]. SIF is a promising proxy of gross primary production (GPP) and has attracted much attention from researchers [410]. The last decade has seen rapid development of atmospheric satellites that made it possible to obtain space-borne SIF products at the global scale [11].

At present, there are no operational sensors specifically designed for SIF monitoring in orbit. Currently, space-borne SIF retrieval mainly relies on the satellites that were originally designed to observe greenhouse gas or trace gas contents and distributions in the atmosphere [11]. The first global SIF map was obtained by Japan’s Greenhouse gases Observing SATellite (GOSAT) [1214], which increased attention on satellite missions for SIF retrieval. Since then, a number of space-borne spectrometers have been employed to obtain global SIF products, such as Orbiting Carbon Observatory-2 (OCO-2) [15, 16], Chinese Carbon Dioxide Observation Satellite Mission (TanSat) [17, 18], Greenhouse gases Observing Satellite-2 (GOSAT-2) [19], and Orbiting Carbon Observatory-3 (OCO-3) [20]. However, the derived SIF datasets only include the near-infrared (NIR) or far-red band because of the limited spectral ranges of these satellites. Satellites with wider spectral ranges have since been used for SIF retrieval at both the far-red and red bands, such as Global Ozone Monitoring Experiment–2 (GOME-2) [21, 22], Scanning Imaging Absorption Spectrometer for Atmospheric Chartography (SCIAMACHY) [2225], and Tropospheric Monitoring Instrument (TROPOMI) [2630]. The FLuorescence EXplorer (FLEX) mission, which was initiated by the European Space Agency in 2015, will be the first satellite to carry a specifically designed SIF spectrometer, but it will not launch until at least 2025 [31, 32]. It is expected to provide a better SIF product that will facilitate a breakthrough in both the spatial and temporal resolutions.

In this context, the upcoming Terrestrial Ecosystem Carbon Inventory Satellite (TECIS-1), is a collaborative mission that was approved by the China Academy of Space Technology and the China National Forestry Administration and is dedicated to the comprehensive monitoring of terrestrial ecosystems. It is scheduled to launch in August 2022, and it will carry four payloads. Among these, the SIF Imaging Spectrometer (SIFIS) is a payload specifically designed for SIF monitoring of the main vegetation types at the global scale [18, 33]. Considering the significance of red SIF for GPP estimation [34], SIFIS is capable of obtaining spectral measurements in the red band. Previous studies have evaluated the performance of SIFIS according to optimized model parameters using a data-driven approach based on preset specifications [33, 35]. However, in situ validation of SIFIS performance has not yet been conducted. SIF makes a small contribution to the total reflected radiation, and SIF retrieval relies on small spectral absorption bands. Thus, accurate radiance measurements are pivotal to accurate SIF retrieval.

We have tried carrying out several in situ experiments to evaluate the performance of SIFIS under real imaging conditions. However, such experiments are very difficult to carry out because they require significant investment in equipment. We successfully conducted one in situ experiment by using SIFIS and the commercial spectrometer QE Pro (Ocean Optic, Inc., Dunedin, FL, USA) to collect canopy radiance spectra of different vegetation types synchronously. This paper presents the results of this experiment. Our aim was (1) to demonstrate and validate the effectiveness of SIFIS on spectral measurement and (2) to compare the SIF retrieval performances of SIFIS and QE Pro.

2. Materials and Methods

2.1. Instruments

TECIS-1 carries four payloads, which includes one SIF spectrometer, SIFIS, with the objective of SIF monitoring for vegetation ecosystems [33]. This push-broom imaging spectrometer is expected to have a spectral range of 664–773 nm with a spectral resolution of 0.3 nm and a spectral sampling interval of 0.1 nm. SIFIS also has a spatial resolution of and swath of ~34 km. The design signal-to-noise ratio (SNR) is greater than 350 at a radiance of 10 mW m−2sr-−1nm−1. The equator crossing time of the satellite is expected to be about 10 : 30 a.m. local time in descending mode, which should let it capture the highest possible values of SIF for a given day.

In this experiment, we used a SIFIS prototype comprising the opto-mechanical main body, signal processor controller, and power box. Other auxiliary equipment including ground examination and map acquisition instruments were also prepared. SIFIS includes the static and push-broom observation modes, each with three gears featuring different angles and spatial resolutions, as given in Table 1.


ModeGearIntegration timeAngle resolution (not digital merging) Original spatial resolution Number of merging pixelsFinal spatial resolution

StaticI70 ms0.021°× 0.037°Spatial: ×8
Frame: ×4
II140 ms0.021°× 0.037°Spatial: ×8
Frame: ×2
III140 ms0.014°× 0.037°Spatial: ×16
Frame: ×2

Push-broomI70 ms0.021°× 0.056°Spatial: ×8
Frame: ×4
II140 ms0.021°× 0.112°Spatial: ×8
Frame: ×2
III140 ms0.014°× 0.112°Spatial: ×16
Frame: ×2

A customized QE Pro spectrometer fitted with an optical fiber after calibration for wavelength and radiometry was employed for spectral measurement of the radiance and SIF retrieval for comparison with SIFIS. QE Pro has a spectral range of 645–805 nm with a spectral resolution of ~0.3 nm and peak SNR of greater than 1000. QE Pro is widely used for ground or tower-based SIF measurements [3438] because of its high-performance specifications.

2.2. Experimental Site

The experiment was carried out over a forest at the Huailai Remote Sensing Test Station (i.e., HL station) in Huailai County, Hebei Province, China, on 23 October 2020. Figure 1(a) displays an overview of the HL station, which is surrounded by various vegetation types and is adjacent to Guanting Reservoir in the north. The HL station was selected owing to the advantageous hardware configuration and various types of land cover available which included crops, forests, water, bare soil, and cement roads. Specifically, the HL station is equipped with an observation platform comprising overhead vehicles and tower cranes that could support observations by the SIFIS prototype above a height of 10 m. However, strong winds resulting from the unique geography and breakdowns of the tower crane caused several experimental attempts on sunny days to fail. Only one experiment was successfully conducted on 23 October 2020, which was relatively sunny with stable weather. Pine and cypress were selected as the vegetation targets, as illustrated in Figures 1(b) and 1(c), because the nearby crops had already been harvested.

2.3. Experimental Scheme

Spectral measurements were made synchronously with SIFIS and QE Pro, which were vertically targeted toward the forest canopy. QE Pro was loaded on SIFIS, and a 5 m-long optical fiber was placed 20 cm from the lens of SIFIS. The fields of view (FOVs) of the lenses of SIFIS and the bare optical fiber were ~3.8° and ~25°, respectively. To reduce differences in their observations, the bare optical fiber of QE Pro was mounted with a lens having an FOV of 3° to acquire data in a wider observation range; we originally intended to hoist the observation platform as high as possible to ~20 m. Moreover, the radiance spectra measured by QE Pro using a bare optical fiber or an optical lens and optical fiber are as shown in Figure S1. However, strong winds prevented the tower crane from increasing the observation platform as desired. Moreover, even on sunny days with little wind, the observation platform still shook when hoisted too high. To maximize the observation range while maintaining stability, the height of the observation platform was eventually set to ~12 m, as illustrated in Figure 1(d).

As presented in Table 1, two observation modes with three different gears were adopted to collect measurements. SIFIS was manually controlled from the ground, and it took several dozens of seconds to 2 min to collect a measurement depending on the stability of the observation platform. In contrast, QE Pro in sequential measurement mode collected one measurement per second during data acquisition. To weaken the influence of noise, a dark correction was first applied to each QE Pro spectral measurement. Then, the measured radiance was taken by averaging the 100 scans with an optimized integration time. During the course of 1 day, observations were collected successfully from ~13 : 30 to ~14 : 30. SIFIS collected 15 images, including six in static mode and nine in push-broom mode. We ignored the spatial differences between the observations of the two instruments and matched simultaneous observations of QE Pro and SIFIS to the observation time.

2.4. SIF Retrieval Method

For a Lambertian surface, the radiance entering a sensor over a vegetation canopy can be expressed as follows [39, 40]: where is the cosine of the solar zenith angle; is the top-of-atmosphere solar irradiance; is the atmospheric path reflectance; is the surface reflectance; is the atmospheric spherical albedo; and are the two-way and upward atmospheric transmittances, respectively; and SIF is the top-of-canopy emitted fluorescence signal.

The data-driven approach of principal component analysis (PCA) was applied to the radiance measurements of SIFIS and QE Pro to retrieve the SIF values. The nonfluorescence contribution of (i.e., first two terms in Eq. (1)) was regarded as the product of low- and high-frequency parts, influenced by atmospheric scattering, surface reflection, and atmospheric transmittance [2123, 25, 41]. We used a low-order polynomial of wavelength and the linear combination of principal component vectors decomposed by PCA to describe low- and high-frequency contributions, respectively. The principal component vectors are derived from the nearly synchronous radiance measurements over nonfluorescent surfaces. In this context, the forward model could be represented as where are the vectors decomposed by PCA, and and are the coefficients of the vectors and polynomial, respectively. is the order of the polynomial, is the number of vectors, is the SIF intensity at a certain wavelength, is the specified spectrum shape of SIF that described by a Gaussian function, and is the upward atmospheric transmittance, which can be firstly estimated using the effective upward transmittance, , and solar and viewing zenith angles based on where is the effective two-way atmospheric transmittance derived from the training dataset by normalizing the solar irradiance radiance spectra using the low-order polynomials, and denote the solar zenith angle and viewing zenith angle, respectively. In the SIF retrieval process for SIFIS and QE Pro spectrometers, 720–758 nm and 682–698 nm fitting windows were employed, and a three or second orders of polynomial and number of PCs of 10 and 15 were chosen at NIR and red bands, respectively. Finally, only , , and are left as the unknown parameters. Thus, the SIF values can be retrieved by solving an ordinary least-square problem. The SIF retrievals of SIFIS and QE Pro spectrometers at 740 nm (NIR band) and 685 nm (red band) were obtained for comparisons.

For the PCA-based SIF retrieval approach, the selection of a representative training dataset is vital for accurate SIF retrieval. Regarding selection of a training dataset for SIFIS, one nonvegetation image was employed to collect the training samples used to generate the vectors for SIF retrieval of SIFIS. In contrast, in the SIF retrieval process of QE Pro, several training samples were manually selected among the nearly simultaneous observations of QE Pro over nonvegetated surfaces during the experiment.

2.5. Quantitative Comparison

To reduce the uncertainties attributed to the mismatch between the observation ranges of the two spectrometers, SIF retrieval was performed by comparing strictly filtered pixels of each SIFIS image with concurrent observations by QE Pro. First, the spatial observation ranges of SIFIS and QE Pro were calculated based on the observation height and FOVs of them, and SIF data were retrieved from corresponding pixels of the SIFIS and QE Pro measurements. Then, the matching pixels of the simultaneous SIFIS and QE Pro measurements were extracted line by line. Finally, a threshold value of 10% was used to filter the matching radiance measurements of the two instruments. For the same observation period and spatial observation range, a pair of radiance measurements whose difference at the NIR band (average radiance from 773 nm to 775 nm) was smaller than 10% was retained for the subsequent comparison. Consequently, six to forty pairs of matching samples were obtained for each image. Figure S2 presents the strictly matched radiance spectra observed by the two instruments for the same image. Much more consistency between each pair of matched radiance spectra was observed after being strictly filtered. The coefficient of determination and root-mean-square error (RMSE) were employed to compare the SIF retrievals of the two spectrometers.

3. Results

3.1. Consistency of Radiance Measurements

After radiometric corrections, raw measurements observed by SIFIS were transferred into the radiance spectrum with the same unit of radiance used by QE Pro. To evaluate the performance of SIFIS in terms of spectral imaging ability and radiometric correction, we randomly searched a pair of synchronous measurements by the two spectrometers for an image observed by SIFIS. Figure 2 displays a pair of matching radiance spectra that was derived from observations over pine trees in static mode. The two spectral curves are highly consistent overall, especially regarding the locations and depth of the absorption lines. Neglecting the differences in FOV, the absolute values of radiance were consistent throughout the spectral range with the largest difference being no more than 5%.

SIFIS clearly had a narrower spectral range than QE Pro. However, it still covers a wider spectral range from the red to NIR bands compared with previous or current satellite sensors on orbit, which will enable satellite-based SIF retrieval in both the red and far-red bands. This will be beneficial for potential applications of SIF remote sensing products, such as improved GPP estimation by using SIF in two bands. Disregarding the absolute values, SIFIS also obtained a smooth spectral curve outside the absorption lines, which demonstrate its reliability with regard to SNR.

3.2. Spatial Distribution of SIF Measurements Observed by SIFIS Instrument

Fifteen images were successfully obtained during the experiment, but only one image includes both vegetation and nonvegetation surfaces. Figure 3 maps the spatial distributions of the SIF retrieval at the far-red and red bands. The normalized difference vegetation index (NDVI) and the near-infrared radiance of vegetation (NIRvR) were obtained in push-broom mode. The spatial distribution of the vegetation was clearly demonstrated by the NDVI results, and SIF retrievals for the different fields showed obvious variations in values. Fields A and B demonstrate the SIF retrieval performance of SIFIS. The NDVI values for fields A and B were higher and lower, respectively, than those from the surrounding pixels. Similar patterns were observed for SIF retrieval at the far-red or red bands, which indicates significant consistency in the spatial distributions of the vegetation. For fields with dense vegetation coverage, there was no obvious variation in NDVI, which was saturated with high values. Thus, NIRvR was also used as an index because previous works have shown that it can be linked with GPP. The differences on spatial distributions of both SIF and NIRvR for dense vegetation fields were clearly observed. The SIF retrieval performance was better at the far-red band than that at the red band, especially for nonvegetated or low-coverage fields. For fields with low NDVI values, some retrieval noise was also observed.

3.3. Comparison of SIF Measurements

To quantify the SIF retrieval performance of SIFIS, we compared the observations of SIFIS and QE Pro. Figure 4 compares the results after the strictly filtered method was applied. SIFIS and QE Pro showed obvious consistency for the SIF retrieval performance at the far-red band (Figure 4(a)) with an value of 0.70 and an RMSE of 0.30 mW m−1 sr−1 nm−1. Scatter was randomly distributed around the 1 : 1 line. However, the SIF retrieval performance at the red band was much less consistent (Figure 4(b)) with a much lower value of . In addition, SIFIS retrieved higher SIF values than QE Pro, which resulted in a slope of 1.21. The worse relationship at the red band may be attributed to the poor quality of SIF retrieval by SIFIS. The error bars of the SIF retrievals by the two spectrometers are also included in Figure 4. Despite the strict filtration of the radiance spectrum, large variations in the SIF values were observed between the synchronous SIFIS and QE Pro measurements. Moreover, the variation of the red band SIF retrieval performance was much greater in several images for SIFIS than for QE Pro.

4. Discussion

4.1. Uncertainties and Limitations Associated with the Experiment

Despite the poor performance in the red band, the radiance measurements and SIF retrieval performances of SIFIS and QE Pro showed close agreement at the far-red band. However, large uncertainties in the comparison method and experimental results remain exist, which limits the comparison accuracy. The major influencing factors are summarized as follows.

First and importantly, the spatial mismatch may have had the largest effect on the comparison accuracy, even at an observation height of ~12 m. The spatial resolution was ~35 mm for SIFIS at this observation height. The FOVs of SIFIS and QE Pro were ~3.8° and ~ 3°, respectively, so the difference in observation range was ~0.2 m. Thus, regardless of the discrepancy in observation locations, this small difference in observation range would cause large uncertainties in the radiance measurements because of the fine spatial resolution of SIFIS. Statistical results show that SIFIS obtained an image in about 18–120 s depending on the stability of the observation platform. As an example, Figure S3 depicts all radiance spectra for one row of the corresponding SIFIS image and simultaneous radiance measurements of QE Pro during the same observation period in static observation mode. The measured radiance spectra showed obvious variations with average coefficient of variance of 21.63% and 40.45%, respectively. Thus, instead of directly averaging SIF retrievals of a whole image of SIFIS, pixels of part of an image were averaged after being strictly filtered for comparison with the synchronous measurements by QE Pro. However, the radiance measurements of the two instruments were still not exactly the same even after being strictly filtered, as shown in Figure S2. This implies large variations in the radiance measurements by either SIFIS or QE Pro, which would inevitably affect the matching accuracy. The optical fiber of QE Pro was placed ~20 cm away from the lens of SIFIS, which also results in some discrepancies in their observation locations and ranges.

Second, despite the fine spatial resolution of SIFIS, the heterogeneity of the surface and leaf characteristics of pine or cypress resulted in a large number of mixed pixels. The mixed pixels had a large influence on the selection of the training dataset, which affected the SIF retrieval accuracy to some extent. In addition, SIF retrievals of these mixed pixels also had large uncertainties. Regardless of SIFIS, QE Pro also included various radiance measurements. Although a simple filter on the spectral quality was conducted during the SIF retrieval process for both SIFIS and QE Pro, the radiance spectra of some mixed pixels were still retained. Especially for the comparison scheme without strict filtering, the mixed pixels could have a large influence.

Third, asynchronous observations between the training and test datasets influenced the comparison between the SIF retrieval performances of the two instruments. Previous works have pointed out that synchronous sampling at spatial and temporal scales is vital for SIF retrieval using data-driven approaches [17, 28, 42]. In this experiment, it was difficult to obtain images quickly and smoothly with SIFIS. In this context, training data for the radiance observation over nonvegetated surfaces were successfully obtained in the final period of the experiment. The largest difference in observation time between the first and last images that include nonvegetated training samples was more than 1.5 hours. Especially for an afternoon in the autumn, the rapid changes in incident radiation greatly affect the SIF retrieval performance based on the in-filling effect of SIF on Fraunhofer lines. Thus, the asynchronous observations of the training and test samples may have also induced uncertainty in the SIF retrieval by SIFIS. In contrast, for SIF retrieval by QE Pro, several training samples were manually selected among the nearly simultaneous observations over nonvegetated surfaces during the experiment. Consequently, the inconsistent observation times of the spectral measurements should have a smaller effect on QE Pro than on SIFIS.

Finally, the limited experimental conditions and environment were another influencing factor. For example, the unstable observation platform caused large measurement uncertainties. In addition, simultaneous observations of two instruments on a calibrated white board were not successfully obtained because of the small size of the white board and instability of the observation platform.

Despite these uncertainties, the radiance measurements and SIF retrievals of the two instruments still showed high consistency in this experiment. Eliminating all sources of uncertainty in a realistic experimental environment would be very difficult and almost impossible. Therefore, we concluded that our experimental results demonstrate the capabilities of SIFIS to some extent.

4.2. Uncertainties Associated with Red Band SIF Retrievals

Although many studies have focused on the SIF retrieval and applications at the far-red or NIR band, several studies have suggested that red band SIF retrieval provides vital supplementary information for various applications [4347]. Red band SIF contains more information from photosystem II (PSII) that is more sensitive to photosynthesis. The important significance of red band SIF to GPP estimation has attracted increasing attention in recent years [34]. Several studies have focused on space-borne SIF retrieval using satellite observations that cover wide spectral ranges, such as GOME-2 and TROPOMI [22, 23]. This led to the decision to design a wider spectral range from the red to NIR bands. However, the comparison results indicate that SIFIS did not perform well at red band SIF retrieval. The SIF retrieval performance was worse at the red band than at the far-red band. There are several reasons that may explain this poor performance.

First, because of the strong reabsorption effect caused by leaves within the canopy, the SIF intensity is much weaker at the red band than at the far-red or NIR band [48, 49]. The spectral range of chlorophyll absorption is overlapped by the red band SIF, which results in a much lower red band SIF at the canopy scale than at the photosystem level. Previous studies have pointed out that less than 5% of red band SIF can escape to the canopy [3]. Second, the spectral shape needs to be accurately modeled for SIF retrieval owing to the in-filling effects on absorption lines [21, 22, 26]. However, the spectral shape of reflectance is more complex at the red band than at the NIR band, which is attributed to the sensitivity of the red band reflectance shape to the physiology and biochemistry of vegetation and their strong effects on absorption at this band [50]. Most studies have used the high-order polynomials to depict the spectral shape of the red band reflectance, which still has large uncertainties [22, 2729]. Third, the depth of the absorption lines is shallower and the fitting window is smaller at the red band than at the NIR band. A slight error of in the estimated reflectance shape will generate a large effect on SIF retrieval at the red band [50]. Despite the lower SIF values at the red band, the reflectance is also smaller at the red band than at the NIR band, which increases the relative contribution of SIF to the total upward radiance. Thus, accurately modeling the spectral shape of the red band reflectance remains a challenge for future investigations to improve the SIF retrieval accuracy at the red band.

5. Conclusions

SIFIS is one of the four payloads onboard the TECIS-1 satellite, which aims to monitor terrestrial resources and ecology and provide measurements for evaluating major national ecological projects. In this study, we presented the first performance evaluation of SIFIS regarding spectral measurements and SIF monitoring based on in situ observations. We employed the typical commercial spectrometer QE Pro for comparison with SIFIS. Nearly synchronous spectral measurements obtained by the two instruments showed good agreement regarding the locations of absorption lines and absolute values of radiance measurements. The spatial distribution of the SIF retrieval by SIFIS in push-broom observation mode was consistent with the NDVI and NIRvR results, which both showed obvious variations in values over vegetated and nonvegetated surfaces. SIFIS and QE Pro showed good agreement for their SIF retrieval performances at the far-red band with an value of 0.70, and an RMSE value of 0.30 mW m−2 sr−1 nm−1. However, the results were not as good at the red band, and SIFIS overestimated the SIF retrieval compared with QE Pro with a lower value of 0.23 and an RMSE of 0.26 mW m−2 sr−1 nm−1. The poor SIF retrieval performance at the red band was attributed to large uncertainties in the methodology. These results are the first to demonstrate the reliability of SIFIS under in situ conditions, and they provide theoretical support for SIF retrieval by TECIS-1 after its upcoming launch.

Data Availability

There is no data associated with this article.

Conflicts of Interest

The contact author has declared that neither they nor their coauthors have any competing interests.

Authors’ Contributions

S. Du proposed the experiment and wrote and revised the paper with L. Liu. X. Liu and J. Chen contributed to the experiment and data analysis.

Acknowledgments

This research was funded by the National Natural Science Foundation of China (41825002).

Supplementary Materials

Figure S1: Radiance spectra measured by QE Pro using a bare optical fiber or an optical lens and optical fiber. Figure S2: matched radiance spectra of an image measured by SIFIS and QE Pro after being strictly filtered. Figure S3: radiance spectra of an image measured synchronously by QE Pro (left) and SIFIS (right). (Supplementary Materials). (Supplementary Materials)

References

  1. L. Gu, J. Han, J. D. Wood, C. Y. Y. Chang, and Y. Sun, “Sun-induced Chl fluorescence and its importance for biophysical modeling of photosynthesis based on light reactions,” The New Phytologist, vol. 223, no. 3, pp. 1179–1191, 2019. View at: Publisher Site | Google Scholar
  2. J. Han, C. Y. Y. Chang, L. Gu et al., “The physiological basis for estimating photosynthesis from Chla fluorescence,” The New Phytologist, vol. 234, no. 4, pp. 1206–1219, 2022. View at: Publisher Site | Google Scholar
  3. A. Porcar-Castell, E. Tyystjarvi, J. Atherton et al., “Linking chlorophyll a fluorescence to photosynthesis for remote sensing applications: mechanisms and challenges,” Journal of Experimental Botany, vol. 65, no. 15, pp. 4065–4095, 2014. View at: Publisher Site | Google Scholar
  4. L. Guanter, Y. Zhang, M. Jung et al., “Global and time-resolved monitoring of crop photosynthesis with chlorophyll fluorescence,” Proceedings of the National Academy of Sciences, vol. 111, no. 14, pp. E1327–E1333, 2014. View at: Publisher Site | Google Scholar
  5. A. Porcar-Castell, Z. Malenovsky, T. Magney et al., “Chlorophyll a fluorescence illuminates a path connecting plant molecular biology to Earth-system science,” Nature plants, vol. 7, no. 8, pp. 998–1009, 2021. View at: Publisher Site | Google Scholar
  6. S. Walther, M. Voigt, T. Thum et al., “Satellite chlorophyll fluorescence measurements reveal large-scale decoupling of photosynthesis and greenness dynamics in boreal evergreen forests,” Global change biology, vol. 22, no. 9, pp. 2979–2996, 2016. View at: Publisher Site | Google Scholar
  7. J. D. Wood, T. J. Griffis, J. M. Baker, C. Frankenberg, M. Verma, and K. Yuen, “Multiscale analyses of solar-induced florescence and gross primary production,” Geophysical Research Letters, vol. 44, no. 1, pp. 533–541, 2017. View at: Publisher Site | Google Scholar
  8. X. Wu, X. Xiao, Y. Zhang et al., “Spatiotemporal consistency of four gross primary production products and solar-induced chlorophyll fluorescence in response to climate extremes across CONUS in 2012,” Journal of Geophysical Research – Biogeosciences, vol. 123, no. 10, pp. 3140–3161, 2018. View at: Publisher Site | Google Scholar
  9. P. Yang and C. van der Tol, “Linking canopy scattering of far-red sun-induced chlorophyll fluorescence with reflectance,” Remote Sensing of Environment, vol. 209, pp. 456–467, 2018. View at: Publisher Site | Google Scholar
  10. Z. Zhang, Y. Zhang, J. M. Chen, W. Ju, M. Migliavacca, and T. S. El-Madany, “Sensitivity of estimated total canopy SIF emission to remotely sensed LAI and BRDF products,” Remote Sensing, vol. 2021, article 9795837, pp. 1–18, 2021. View at: Publisher Site | Google Scholar
  11. G. H. Mohammed, R. Colombo, E. M. Middleton et al., “Remote sensing of solar-induced chlorophyll fluorescence (SIF) in vegetation: 50 years of progress,” Remote Sensing of Environment, vol. 231, article 111177, 2019. View at: Publisher Site | Google Scholar
  12. C. Frankenberg, J. B. Fisher, J. Worden et al., “New global observations of the terrestrial carbon cycle from GOSAT: Patterns of plant fluorescence with gross primary productivity,” Geophysical Research Letters, vol. 38, no. 17, 2011. View at: Publisher Site | Google Scholar
  13. L. Guanter, C. Frankenberg, A. Dudhia et al., “Retrieval and global assessment of terrestrial chlorophyll fluorescence from GOSAT space measurements,” Remote Sensing of Environment, vol. 121, pp. 236–251, 2012. View at: Publisher Site | Google Scholar
  14. J. Joiner, Y. Yoshida, A. P. Vasilkov, and E. M. Middleton, “First observations of global and seasonal terrestrial chlorophyll fluorescence from space,” Biogeosciences, vol. 8, no. 3, pp. 637–651, 2011. View at: Publisher Site | Google Scholar
  15. C. Frankenberg, C. O'Dell, J. Berry et al., “Prospects for chlorophyll fluorescence remote sensing from the orbiting carbon Observatory-2,” Remote Sensing of Environment, vol. 147, pp. 1–12, 2014. View at: Publisher Site | Google Scholar
  16. Y. Sun, C. Frankenberg, M. Jung et al., “Overview of solar-induced chlorophyll fluorescence (SIF) from the orbiting carbon Observatory-2: retrieval, cross-mission comparison, and global monitoring for GPP,” Remote Sensing of Environment, vol. 209, pp. 808–823, 2018. View at: Publisher Site | Google Scholar
  17. S. Du, L. Liu, X. Liu et al., “Retrieval of global terrestrial solar-induced chlorophyll fluorescence from TanSat satellite,” Scientific Bulletin, vol. 63, no. 22, pp. 1502–1512, 2018. View at: Publisher Site | Google Scholar
  18. S. Du, L. Liu, X. Liu, and J. Chen, “First investigation of the relationship between solar-induced chlorophyll fluorescence observed by TanSat and gross primary productivity,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 11892–11902, 2021. View at: Publisher Site | Google Scholar
  19. H. Oshio, Y. Yoshida, and T. Matsunaga, “Preliminary analysis toward GOSAT-2 new products, proxy-based XCH4 and SIF, using GOSAT data,” Tech. Rep., AGU Fall Meeting Abstracts, 2017. View at: Google Scholar
  20. A. Eldering, T. E. Taylor, C. W. O'Dell, and R. Pavlick, “The OCO-3 mission: measurement objectives and expected performance based on 1 year of simulated data,” Atmospheric Measurement Techniques, vol. 12, no. 4, pp. 2341–2370, 2019. View at: Publisher Site | Google Scholar
  21. J. Joiner, L. Guanter, R. Lindstrot et al., “Global monitoring of terrestrial chlorophyll fluorescence from moderate spectral resolution near-infrared satellite measurements: methodology, simulations, and application to GOME-2,” Atmospheric Measurement Techniques, vol. 6, no. 10, pp. 2803–2823, 2013. View at: Publisher Site | Google Scholar
  22. J. Joiner, Y. Yoshida, L. Guanter, and E. M. Middleton, “New methods for the retrieval of chlorophyll red fluorescence from hyperspectral satellite instruments: simulations and application to GOME-2 and SCIAMACHY,” Atmospheric Measurement Techniques, vol. 9, no. 8, pp. 3939–3967, 2016. View at: Publisher Site | Google Scholar
  23. P. Köhler, L. Guanter, and J. Joiner, “A linear method for the retrieval of sun-induced chlorophyll fluorescence from GOME-2 and SCIAMACHY data,” Atmospheric Measurement Techniques, vol. 8, no. 6, pp. 2589–2608, 2015. View at: Publisher Site | Google Scholar
  24. J. Joiner, Y. Yoshida, A. P. Vasilkov et al., “Filling-in of near-infrared solar lines by terrestrial fluorescence and other geophysical effects: simulations and space-based observations from SCIAMACHY and GOSAT,” Atmospheric Measurement Techniques Discussions, vol. 5, no. 4, pp. 809–829, 2012. View at: Publisher Site | Google Scholar
  25. E. Van Schaik, M. L. Kooreman, P. Stammes et al., “Improved SIFTER v2 algorithm for long-term GOME-2A satellite retrievals of fluorescence with a correction for instrument degradation,” Atmospheric Measurement Techniques, vol. 13, no. 8, pp. 4295–4315, 2020. View at: Publisher Site | Google Scholar
  26. L. Guanter, I. Aben, P. Tol et al., “Potential of the TROPOspheric monitoring instrument (TROPOMI) onboard the Sentinel-5 precursor for the monitoring of terrestrial chlorophyll fluorescence,” Atmospheric Measurement Techniques, vol. 8, no. 3, pp. 1337–1352, 2015. View at: Publisher Site | Google Scholar
  27. L. Guanter, C. Bacour, A. Schneider et al., “The TROPOSIF global sun-induced fluorescence dataset from the sentinel-5P TROPOMI mission,” Earth System Science Data, vol. 13, no. 11, pp. 5423–5440, 2021. View at: Publisher Site | Google Scholar
  28. P. Köhler, C. Frankenberg, T. S. Magney, L. Guanter, J. Joiner, and J. Landgraf, “Global retrievals of solar-induced chlorophyll fluorescence with TROPOMI: first results and intersensor comparison to OCO-2,” Geophysical Research Letters, vol. 45, pp. 10456–10463, 2018. View at: Publisher Site | Google Scholar
  29. P. Köhler, M. J. Behrenfeld, J. Landgraf, J. Joiner, T. S. Magney, and C. Frankenberg, “Global retrievals of solar-induced chlorophyll fluorescence at red wavelengths with TROPOMI,” Geophysical Research Letters, vol. 47, no. 15, 2020. View at: Publisher Site | Google Scholar
  30. F. Zhao, W. Ma, P. Köhler et al., “Retrieval of red solar-induced chlorophyll fluorescence with TROPOMI on the Sentinel-5 precursor Mission,” IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1–14, 2022. View at: Publisher Site | Google Scholar
  31. P. Coppo, A. Taiti, L. Pettinato, M. Francois, M. Taccola, and M. Drusch, “Fluorescence imaging spectrometer (FLORIS) for ESA FLEX Mission,” Remote Sensing, vol. 9, no. 7, p. 649, 2017. View at: Publisher Site | Google Scholar
  32. M. Drusch, J. Moreno, U. Del Bello et al., “The FLuorescence EXplorer mission concept - ESA's earth Explorer 8,” IEEE Transactions on Geoscience and Remote Sensing, vol. 55, no. 3, pp. 1273–1284, 2017. View at: Publisher Site | Google Scholar
  33. S. Du, L. Liu, X. Liu, X. Zhang, X. Gao, and W. Wang, “The Solar-Induced Chlorophyll Fluorescence Imaging Spectrometer (SIFIS) Onboard the First Terrestrial Ecosystem Carbon Inventory Satellite (TECIS-1): Specifications and Prospects,” Sensors, vol. 20, no. 3, 2020. View at: Publisher Site | Google Scholar
  34. X. Liu, L. Liu, J. Hu, J. Guo, and S. Du, “Improving the potential of red SIF for estimating GPP by downscaling from the canopy level to the photosystem level,” Agricultural and Forest Meteorology, vol. 281, article 107846, 2020. View at: Publisher Site | Google Scholar
  35. J. Chen, X. Liu, S. Du, Y. Ma, and L. Liu, “Integrating SIF and Clearness Index to Improve Maize GPP Estimation Using Continuous Tower-Based Observations,” Sensors, vol. 20, no. 9, 2020. View at: Publisher Site | Google Scholar
  36. X. Liu, Z. Liu, L. Liu et al., “Modelling the influence of incident radiation on the SIF-based GPP estimation for maize,” Agricultural and Forest Meteorology, vol. 307, article 108522, 2021. View at: Publisher Site | Google Scholar
  37. T. Julitta, L. A. Corp, M. Rossini et al., “Comparison of sun-induced chlorophyll fluorescence estimates obtained from four portable field spectroradiometers,” Remote Sensing, vol. 8, no. 2, p. 122, 2016. View at: Publisher Site | Google Scholar
  38. N. Nakashima, T. Kato, T. Morozumi et al., “Area-ratio Fraunhofer line depth (aFLD) method approach to estimate solar-induced chlorophyll fluorescence in low spectral resolution spectra in a cool-temperate deciduous broadleaf forest,” Journal of Plant Research, vol. 134, no. 4, pp. 713–728, 2021. View at: Publisher Site | Google Scholar
  39. L. Guanter, L. Alonso, L. Gómez-Chova et al., “Developments for vegetation fluorescence retrieval from spaceborne high-resolution spectrometry in the O2-A and O2-B absorption bands,” Journal of Geophysical Research, vol. 115, no. D19, 2010. View at: Publisher Site | Google Scholar
  40. X. Liu and L. Liu, “Assessing band sensitivity to atmospheric radiation transfer for space-based retrieval of solar-induced chlorophyll fluorescence,” Remote Sensing, vol. 6, no. 11, pp. 10656–10675, 2014. View at: Publisher Site | Google Scholar
  41. C. Zou, S. Du, X. Liu, L. Liu, Y. Wang, and Z. Li, “Optimizing the Empirical Parameters of the Data-Driven Algorithm for SIF Retrieval for SIFIS Onboard TECIS-1 Satellite,” Sensors, vol. 21, no. 10, 2021. View at: Publisher Site | Google Scholar
  42. L. Guanter, M. Rossini, R. Colombo et al., “Using field spectroscopy to assess the potential of statistical approaches for the retrieval of sun-induced chlorophyll fluorescence from ground and space,” Remote Sens. of Environ, vol. 133, pp. 52–61, 2013. View at: Publisher Site | Google Scholar
  43. G. Agati, P. Mazzinghi, F. Fusi, and I. Ambrosini, “The F685/F730 chlorophyll fluorescence ratio as a tool in plant physiology: response to physiological and environmental factors,” Journal of Plant Physiology, vol. 145, no. 3, pp. 228–238, 1995. View at: Publisher Site | Google Scholar
  44. H. K. Lichtenthaler and F. Babani, “Detection of photosynthetic activity and water stressby imaging the red chlorophyll fluorescence,” Plant Physiology and Biochemistry, vol. 38, no. 11, pp. 889–895, 2000. View at: Publisher Site | Google Scholar
  45. L. Liu, X. Liu, J. Hu, and L. Guan, “Assessing the wavelength-dependent ability of solar-induced chlorophyll fluorescence to estimate the GPP of winter wheat at the canopy level,” International Journal of Remote Sensing, vol. 38, no. 15, pp. 4396–4417, 2017. View at: Publisher Site | Google Scholar
  46. M. Rossini, L. Nedbal, L. Guanter et al., “Red and far red sun-induced chlorophyll fluorescence as a measure of plant photosynthesis,” Geophysical Research Letters, vol. 42, no. 6, pp. 1632–1639, 2015. View at: Publisher Site | Google Scholar
  47. P. J. Zarco-Tejada, J. C. Pushnik, S. Dobrowski, and S. L. Ustin, “Steady-state chlorophyll a fluorescence detection from canopy derivative reflectance and double-peak red-edge effects,” Remote Sensing of Environment, vol. 84, no. 2, pp. 283–294, 2003. View at: Publisher Site | Google Scholar
  48. G. Agati, F. Fusi, P. Mazzinghi, and M. L. di Paola, “A simple approach to the evaluation of the reabsorption of chlorophyll fluorescence spectra in intact leaves,” Journal of Photochemistry and Photobiology. B, vol. 17, no. 2, pp. 163–171, 1993. View at: Publisher Site | Google Scholar
  49. J. M. Romero, G. B. Cordon, and M. G. Lagorio, “Modeling re-absorption of fluorescence from the leaf to the canopy level,” Remote Sensing of Environment, vol. 204, pp. 138–146, 2018. View at: Publisher Site | Google Scholar
  50. L. Liu, X. Liu, and J. Hu, “Effects of spectral resolution and SNR on the vegetation solar-induced fluorescence retrieval using FLD-based methods at canopy level,” European Journal of Remote Sensing, vol. 48, no. 1, pp. 743–762, 2015. View at: Publisher Site | Google Scholar

Copyright © 2022 Shanshan Du et al. Exclusive Licensee Aerospace Information Research Institute, Chinese Academy of Sciences. Distributed under a Creative Commons Attribution License (CC BY 4.0).

 PDF Download Citation Citation
Views17
Downloads17
Altmetric Score
Citations