BOREAS TE-18 GeoSail Canopy Reflectance Model Summary: The SAIL (Scattering from Arbitrarily Inclined Leaves) model was combined with the Jasinski geometric model to simulate canopy spectral reflectance and absorption of photosynthetically active radiation for discontinuous canopies. This model is called the GeoSail model. Tree shapes are described by cylinders or cones distributed over a plane. Spectral reflectance and transmittance of trees are calculated from the SAIL model to determine the reflectance of the three components used in the geometric model: illuminated canopy, illuminated background, shadowed canopy, and shadowed background. The model code is Fortran, sample input and output data are provided in ASCII text files. Table of Contents * 1 Model Overview * 2 Investigator(s) * 3 Model Theory * 4 Equipment * 5 Data Acquisition Methods * 6 Observations * 7 Data Description * 8 Data Organization * 9 Data Manipulations * 10 Errors * 11 Notes * 12 Application of the Model * 13 Future Modifications and Plans * 14 Software * 15 Data Access * 16 Output Products and Availability * 17 References * 18 Glossary of Terms * 19 List of Acronyms * 20 Document Information 1. Model Overview 1.1 Model Identification BOREAS TE-18 GeoSail Canopy Reflectance Model 1.2 Model Introduction The GeoSail model is made up of two parts: the SAIL that calculates within crown radiative transfer, and the Geo part that determines amount of shadowing in a scene. The program consists of a driver that reads and writes the model input and output and calls subroutines that make up the model. This code has been compiled on Vax and Macintosh computers without problems. 1.3 Objective/Purpose The purpose of the GeoSail model was to develop a relatively simple model that would be able to describe the reflectance characteristics of boreal forests based on ground-based measurements of the canopy characteristics. 1.4 Summary of Parameters Model inputs include: ? Canopy components ? Quantity of each component ? Inclination angle distribution ? Spectral reflectance and transmission for each wave band ? Leaf area index (LAI) within the tree crown ? Spectral reflectance of the background for each wave band ? solar zenith angle ? shape of the crowns (either cylinder or cone) ? height to width ratio of the crown ? crown coverage Model outputs include: ? Nadir view reflectance of the scene for each waveband ? Total scene LAI ? Fraction of photosynthetically active radiation (PAR) absorbed by the canopy (Fpar). Two sample input and output data files are provided. Both calculate visible and near infrared reflectance and Fpar, one is for aspen stands and the other is for spruce stands. 1.5 Discussion GeoSail uses the SAIL model to calculate canopy spectral reflectance and transmittance. Shadowed background reflectance is the product of the background reflectance and the transmission from SAIL. The component reflectances: illuminated canopy, illuminated background, shadowed canopy, and shadowed background are then used as inputs into a geometric model. The geometric model determines the fraction of each of the components in a scene, given a crown shape, the canopy coverage, and the sun angle. In this model all tree crowns are assumed to be identical and do not overshadow each other. 1.6 Related Models The SAIL model was developed by Verhoef (1984), the code is a modified version of the code written by Alexander (1983). The SAIL model is in subroutine sail, the sail subroutine calls a second subroutine mmult (a matrix multiplication subroutine). The geometric part of the model is based on Jasinski's geometric model (Jasinski 1990, Jasinski and Eagleson 1989 and 1990). A complete description of GeoSail is in Huemmrich (1995). There are two GeoSail subroutines: geocyli and geocone, to describe tree crowns that are shaped like cylinders or cones. These subroutines use crown reflectance and transmittance values passed from the sail subroutine. 2. Investigator 2.1 Investigator Name and Title K. Fred Huemmrich Assistant Research Scientist 2.2 Title of Investigation BOREAS Staff Science 2.3 Contact Information Contact 1 --------- K. Fred Huemmrich University of Maryland NASA GSFC Greenbelt, MD (301) 286-4862 (301) 286-0239 (fax) Karl.Huemmrich@gsfc.nasa.gov 3. Model Theory The GeoSail model combines a geometric model, which calculates the amount of shadowed and illuminated components in a scene, with the SAIL model, which calculates the reflectance and transmittance of the tree crowns. Scene reflectance is determined in the model by calculating an area-weighted average of three landscape components: illuminated canopy, illuminated background, and shadowed background. The simplicity of this model is a result of the assumption that single values for the reflectance and transmittance of light by canopy clumps are enough to provide a reasonable overall description of scene reflectance and absorption. The geometric part of GeoSail uses the model developed by Jasinski for the limiting case where the shadows cast by clumps of vegetation are very small relative to the size of the area observed (Jasinski and Eagleson 1989; Jasinski 1990; Jasinski and Eagleson 1990). The Jasinski model consists of a scene made up of geometric solids scattered over a plane with a Possion distribution. The solids are identical in size and shape and cast shadows on the background plane, but do not overshadow each other. The fraction of a scene that is shadowed is determined by S = 1 - C - (1 - C)(?+1) (1) where S is the fraction of shadowed background and C is the fraction covered by the solids, i.e. the canopy coverage. The parameter ? is the ratio of canopy cover to shadow area for a single geometric solid. Thus the shadowed fraction will vary with differences in the geometry of the canopy components and the solar zenith angle. Trees display a wide variety of shapes. In general, tree shape can be described with three parameters: size, ratio of height to width, and shape. The height to width ratio should adjust with conditions, as the tree increases in height to reach the canopy and in width to occupy space in the canopy and to intercept light. The balance between growth in height and width depends on the growth strategies of the tree and the surrounding trees, the environment where the tree is growing, and that per unit length horizontal branches are more expensive than vertical growth. The shape of a tree determines the total amount of light it can intercept and defines limits to leaf placement. Many shapes can be used to describe the tree forms, for example spheres, cones, and ellipsoids have all been used in past modeling work. In Jasinski’s model the ? term makes it easy to describe different tree shapes. For cylindrical solids ? is calculated using the height to width ratio of the cylinder, R, and the solar zenith angle, ? ? = R tan(?). (2a) ? can also be calculated for cones as: ? = (tan(?) - ?)/?. (2b) Where ? is determined from the aspect angle of the cone, ?, and the solar zenith angle as ? = Arccos(tan(?)/tan(?)). (3) The fraction of the area that is illuminated background, B, can then be calculated by B = 1 - C - S. (4) To calculate the scene reflectance the reflectance of each component in the scene is weighted by its fractional area and summed: ?t = C ?c + S ?s + B ?b (5) Where rt is the total scene reflectance and ?c, ?s, and ?b are, respectively, canopy, shadow, and background reflectances in a specific waveband. Equation 5 assumes that there is no interaction between the components, such as mutual shading. The canopy and shadow reflectances are obtained from SAIL model results. Canopy reflectance, ?c, is the SAIL model reflectance with LAI set at the value for the LAI for a single tree. The illuminated background reflectance, ?b, is the same background reflectance used in the SAIL model. Shadows are areas of lower illumination due to light absorption by the canopy, so the shadow reflectance, ?s, is the product of the transmittance through the canopy, calculated by the SAIL model, and the background reflectance. This formulation of scene reflectance allows for multiple scattering within tree canopies but assumes no multiple scattering between trees. If the trees are described as cones and the aspect angle of the cone is less than the solar zenith angle, then a portion of the cone is shadowed. From a nadir view the fraction of shadowed canopy (Cs) is Cs = ?/?. (6) The shadowed canopy is assigned its own reflectance value, adding another term to equation 5. GeoSail calculates the fraction of absorbed photosynthetically active radiation (Fapar) differently from the SAIL model. In the SAIL calculations of Fapar the PAR fluxes into and out of the canopy are determined and the absorbed photosynthetically active radiation (APAR) is calculated as the difference between the PAR into and out of the canopy (Goward and Huemmrich 1992). In GeoSail the method used to calculate Fapar follows the approach of Bégué (1991). The fraction of absorbed PAR is calculated by, Fapar = (1-?c){(C+S) + (B ?b ID ) + (S ?s ID )} + {C (?s -?HC )} (7) where ?c is the transmittance of PAR through the canopy and ?HC is the hemispheric reflectance from the top of the canopy. Values from both of these variables come from the SAIL model. ID is the diffuse interception, that is the fraction of non-directional light which is absorbed by the canopy clumps. This is calculated by integrating the interception of light by the canopy for all possible solar zenith angles. Diffuse interception is given by, p/2 ID =?(C + S) d?. (8) ?=0 It is the amount of light intercepted by the canopy clumps as the solar zenith angle goes from 0 to 90 degrees. The components of equation 7 can be explained as follows: the term C+S is the fraction of the incoming light that intercepts a canopy clump; the term B ?b ID is the radiation which directly hits the background and is reflected back into the canopy, assuming the background is a Lambertian reflector; S ?s ID is the light reflected from the shadowed background that is reflected back into the canopy; and C (?s -?HC ) is the light reflected from the background under the vegetation clump into the canopy minus the light reflected from the top of the vegetation clumps. 4. Equipment: Not Applicable. 5. Data Acquisition Methods Leaf and twig spectral reflectance and transmittance values come from the BOREAS Terrestrial Ecology (TE)-12 group: Radiation and Gas Exchange of Canopy Elements in a Boreal Forest. Background reflectance values were collected by the BOREAS Remote Sensing Science (RSS)-19 group: Variation in Radiometric Properties of the Boreal Forest Landscape as a Function of the Ecosystem Dynamics. The high spectral resolution data were averaged to broad bands. Crown LAI values came from TE-6: Measurement and Scaling of Carbon Budgets for Contrasting Boreal Forest Sites. Estimates of canopy cover came from tables in the 1994 BOREAS Experiment Plan (EXPLAN94). 6. Observations 6.1 Data Notes None. 7. Data Description 7.1 Spatial Characteristics 7.1.1 Spatial Coverage The GeoSail model does not have a specific spatial scale. The model assumptions, however, are based on viewing a scene where the size of the shadows cast by the trees are small relative to the size of the pixel. 7.1.2 Spatial Coverage Map Not Applicable. 7.1.3 Spatial Resolution Not Applicable. 7.1.4 Projection Not Applicable. 7.1.5 Grid Description Not Applicable. 7.2 Temporal Characteristics 7.2.1 Temporal Coverage Not Applicable. 7.2.2 Temporal Coverage Map Not Applicable. 7.2.3 Temporal Resolution Not Applicable. 7.3 Data Characteristics 7.3.1 Input Parameter/Variable The GeoSail program asks for the input file name, the input file should be a text file structured as shown below. The following sample input file is in the file Aspen_In.DAT: Aspen_Out.DAT 2 1 2 ASPEN WITH 2% TWIGS, 98% GREEN 5. 15. 25. 35. 45. 55. 65. 75. 85. .015211 .04517 .07376 .10010 .12341 .14297 .15818 .16858 .17387 0.2205 0.2073 0.1825 0.1491 0.1111 0.0731 0.0397 0.0149 0.0017 0.070 0.505 0.032 0.407 0.245 0.677 0.000 0.000 0.132 0.225 42.27 4.90 0.1 CYLI 3.5 10 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 The first line is the name of the output file, which can contain up to 25 characters. The second line gives input values for the model, there should be three integers separated by spaces. These input values are as follows: 1. number of different wavelength bands to calculate (up to 10) 2. number of canopy layers (up to 15) 3. number of different types of components in the canopy (up to 15) (this example shows 2 wavelength bands, 1 canopy layer, and 2 canopy components) The third line a comment line to describe the run, up to 50 characters The fourth line is the midpoint angles for the nine leaf inclination angle bins. These should be floating point numbers separated by spaces The next lines are the fraction of the leaves that fall into each leaf angle bin; there will be one line for each canopy component. (In this example the first line are the values for a spherical leaf angle distribution, and the second line is a planophile distribution) The next two lines are the reflectances and transmittances of the first canopy component. The first line has the reflectances of the canopy component for each of the wavelengths. The second line has the transmittances. As there is more than one component in the canopy being modeled there is other pair of lines following the first pair describing the spectral reflectance and transmittance of the second component. This pattern would repeat for each additional canopy component. (In this example the first line is aspen leaf reflectance in red and near infrared bands, the second line is the leaf transmittance in the same two wavelength bands, the third line is aspen branch reflectance, and the fourth is the branch transmittance [set to zero]) After the component reflectance and transmittance lines is the background reflectance values for each wavelength band. (In this example the background is leaf litter) The next line is the solar zenith angle for the simulation. The following line has the LAI of each of the components. These are the LAI values to describe an individual tree crown. If there were multiple layers specified, the LAI for the first component of the first layer is the first number. This is followed by the LAI for the first component of the second layer, continuing through all the layers, then repeating the pattern for the second component through all the layers, and so on. All of the values would be separated by spaces in the same line. (In this example the leaves have an LAI of 4.9 and the branches have an LAI of 0.1, for a crown plant area index [PAI] of 5) All of the inputs up to this point go into the SAIL model, which then calculates the crown spectral reflectance and transmittance. The next line is a four-letter code to describe the crown shapes as cylinders (CYLI) or cones (CONE). The next line is the height to width ratio of the crowns. (In this case the trees are described as cylinders, the height of which is 3.5 times the diameter.) The next line gives the number of different canopy coverages to calculate reflectances. The next lines (the number of them is given above) each have a different value for the fraction of canopy cover, that is the fraction of the scene covered by the cones or cylinders. (In this case there are 10 different coverages to be calculated, starting with 0.1 through 1.0) To the output file, the program writes out all of the input values. From the SAIL model the nadir viewed reflectance, the transmittance, and the hemispheric reflectance for each wavelength band are also written to the output file. From the GeoSail model the total LAI, that is the product of the crown PAI and the fraction canopy cover, the nadir view reflectance, and the instantaneous fraction of radiation absorbed for each band are output. In the case where the input parameters are for the PAR wavelengths, the fraction of radiation absorbed is Fapar. The file SPRUIN.DAT is a sample GeoSail input file for calculating red and near infrared reflectance for spruce trees. The file ASPNIN.DAT is a sample GeoSail input file for calculating red and near infrared reflectance for aspen trees. 7.3.2 Variable Description/Definition See Section 7.3.1 7.3.3 Unit of Measurement The measurement units for the variables contained in the input data files are: Variable Description Units ---------------------------------------------------------------- ----------------- Output file name [none] Number of different wavelength bands to calculate [count] Number of canopy layers [count] Number of different types of components in the canopy [count] Comments [none] Midpoint of leaf inclination angle bins [degrees] Fraction of the leaves which fall into each leaf angle bin [unitless] Canopy component spectral reflectance [unitless] Canopy component spectral transmittance [unitless] Background spectral reflectance [unitless] Solar zenith angle [degrees] LAI of canopy component [unitless] Code for crown shape [none] Crown height to width ratio [unitless] Number of different canopy cover values [count] Canopy cover [unitless] 7.3.4 Data Source The source for the variables contained in the input data files are: Variable Description Data Source ---------------------------------------------------------------- ----------------- Output file name [investigator] Number of different wavelength bands to calculate [investigator] Number of canopy layers [investigator] Number of different types of components in the canopy [investigator] Comments [investigator] Midpoint of leaf inclination angle bins [investigator] Fraction of the leaves which fall into each leaf angle bin [investigator] Canopy component spectral reflectance [TE-12] Canopy component spectral transmittance [TE-12] Background spectral reflectance [RSS-19] Solar zenith angle [investigator] LAI of canopy component [TE-06] Code for crown shape [investigator] Crown height to width ratio [investigator] Number of different canopy cover values [investigator] Canopy cover [EXPLAN94] 7.3.5 Data Range The range of values for the variables contained in the input data files are: Minimum Maximum Data Data Variable Description Value Value --------------------------------------------------------- ------------ ---------- Output file name N/A N/A Number of different wavelength bands to calculate 1 10 Number of canopy layers 1 15 Number of different types of components in the canopy 1 15 Comments N/A N/A Midpoint of leaf inclination angle bins 1 89 Fraction of the leaves which fall into each leaf angle bin 0 1 Canopy component spectral reflectance 0 1 Canopy component spectral transmittance 0 1 Background spectral reflectance 0 1 Solar zenith angle 0 89 LAI of canopy component 0 999 Code for crown shape CONE CYLI Crown height to width ratio 0 999 Number of different canopy cover values 1 999 Canopy cover 0 1 Note: the value 999 means the maximum size is arbitrary. 7.4.1 Output Parameter/Variable The output file produced by GeoSail from the sample input file Aspen_In.Dat (Section 7.3.1) is shown below and is stored in the file Aspen_Out.Dat. OUTPUT FROM: ASPEN_IN.DAT ASPEN WITH 2% TWIGS, 98% GREEN LEAF INCLINATION ANGLES 5.0 15.0 25.0 35.0 45.0 55.0 65.0 75.0 85.0 0.0152 0.0452 0.0738 0.1001 0.1234 0.1430 0.1582 0.1686 0.1739 0.2205 0.2073 0.1825 0.1491 0.1111 0.0731 0.0397 0.0149 0.0017 COMPONENT OPTICAL PROPS COMPONENT BAND REFL TRANS 1 1 0.0700 0.0320 1 2 0.5050 0.4070 2 1 0.2450 0.0000 2 2 0.6770 0.0000 SOIL REFLECTANCE BAND SOIL REFL 1 0.1320 2 0.2250 SOLAR ZENITH ANGLE: 42.27 SAIL MODEL OUTPUT TOTAL CROWN LAI= 5.000 BAND REFL TRANS HEMI REFL 1 0.0253 0.0360 0.0275 2 0.4175 0.2313 0.4739 GEOSAIL OUTPUT CYLINDRICAL CROWNS HEIGHT/WIDTH= 3.500 CANOPY COVER TOTAL LAI BAND REFL FRAC ABS 0.100 0.500 1 0.0887 0.3793 0.100 0.500 2 0.1999 0.2875 0.200 1.000 1 0.0589 0.6132 0.200 1.000 2 0.1932 0.4355 0.300 1.500 1 0.0396 0.7632 0.300 1.500 2 0.2006 0.5123 0.400 2.000 1 0.0280 0.8551 0.400 2.000 2 0.2186 0.5417 0.500 2.500 1 0.0220 0.9074 0.500 2.500 2 0.2443 0.5393 0.600 3.000 1 0.0198 0.9335 0.600 3.000 2 0.2750 0.5162 0.700 3.500 1 0.0200 0.9438 0.700 3.500 2 0.3090 0.4806 0.800 4.000 1 0.0214 0.9457 0.800 4.000 2 0.3446 0.4382 0.900 4.500 1 0.0233 0.9439 0.900 4.500 2 0.3809 0.3930 1.000 5.000 1 0.0253 0.9412 1.000 5.000 2 0.4175 0.3469 Output file description is as follows. The first line is the comments line from the input file. Under the heading "LEAF INCLINATION ANGLES", the first line is the midpoint of each of the leaf inclination angle bins. The next lines are the fraction of the leaves which fall into each leaf angle bin, there will be one line for each canopy component. Under the heading "COMPONENT OPTICAL PROPS", are the input values of the component spectral reflectance and transmittance for each wavelength band. The columns are the component number, the wavelength band number, the spectral reflectance, and the spectral transmittance. Under the heading "SOIL REFLECTANCE", are the input values of the background spectral reflectance for each wavelength band. The columns are the wavelength band number and the spectral reflectance for that band. Next the input solar zenith angle is given. The output of the SAIL model is listed under the heading "SAIL MODEL OUTPUT." The total crown LAI is the sum of the input LAI values for all of the different components. From the SAIL model the nadir viewed reflectance, the transmittance, and the hemispheric reflectance for each wavelength band are listed. The GeoSail output is listed under the heading "GEOSAIL OUTPUT." The type of canopy crown shapes used in the calculation is listed along with the input height to width ratio. The following table lists the total LAI, that is the product of the total crown LAI and the fraction canopy cover, the wavelength band number, the nadir view reflectance, and the instantaneous fraction of radiation absorbed for that wavelength band. In the case where the wavelength band is for the PAR wavelengths, the fraction of radiation absorbed is Fapar. The file SPRUOUT.DAT is a sample GeoSail output file, the result of running the input file SPRUIN.DAT. SPRUOUT.DAT contains calculated red and near infrared reflectance for spruce trees. The file ASPNOUT.DAT is a sample GeoSail output file, the result of running the input file ASPNIN.DAT. ASPNOUT.DAT contains calculated red and near infrared reflectance for aspen trees. 7.4.2 Variable Description/Definition See section 7.4.1. 7.4.3 Unit of Measurement The measurement units for the variables contained in the output data files are: Variable Description Units ---------------------------------------------------------------- ----------------- Input file information [none] Comments [none] Midpoint of leaf inclination angle bins [degrees] Fraction of the leaves which fall into each leaf angle bin [unitless] Wavelength band number [count] Canopy component number [count] Canopy component spectral reflectance [unitless] Canopy component spectral transmittance [unitless] Background spectral reflectance [unitless] Solar zenith angle [degrees] Total crown LAI [unitless] SAIL canopy nadir reflectance [unitless] SAIL canopy transmittance [unitless] SAIL canopy hemispherical reflectance [unitless] Crown shape [none] Crown height to width ratio [unitless] Canopy cover [unitless] Total LAI [unitless] GeoSail nadir reflectance [unitless] GeoSail instantaneous absorption [unitless] 7.4.4 Data Source The data source for the variables contained in the output data files are: Variable Description Data Source ---------------------------------------------------------------- ----------------- Input file information [input file] Comments [input file] Midpoint of leaf inclination angle bins [input file] Fraction of the leaves which fall into each leaf angle bin [input file] Wavelength band number [input file] Canopy component number [input file] Canopy component spectral reflectance [input file] Canopy component spectral transmittance [input file] Background spectral reflectance [input file] Solar zenith angle [input file] Total crown LAI [calculated] SAIL canopy nadir reflectance [SAIL model] SAIL canopy transmittance [SAIL model] SAIL canopy hemispherical reflectance [SAIL model] Crown shape [input file] Crown height to width ratio [input file] Canopy cover [input file] Total LAI [calculated] GeoSail nadir reflectance [GeoSail] GeoSail instantaneous absorption [GeoSail] 7.4.5 Data Range The range of values for the variables contained in the output data files are: Minimum Maximum Data Data Variable Description Value Value --------------------------------------------------------- ------------ ---------- Input file information N/A N/A Comments N/A N/A Midpoint of leaf inclination angle bins 1 89 Fraction of the leaves in each leaf angle bin 0 1 Wavelength band number 1 10 Canopy component number 1 15 Canopy component spectral reflectance 0 1 Canopy component spectral transmittance 0 1 Background spectral reflectance 0 1 Solar zenith angle 0 89 Total crown LAI 0 999 SAIL canopy nadir reflectance 0 1 SAIL canopy transmittance 0 1 SAIL canopy hemispherical reflectance 0 1 Crown shape CONE CYLINDER Crown height to width ratio 0 999 Canopy cover 0 1 Total LAI 0 999 GeoSail nadir reflectance 0 1 GeoSail instantaneous absorption 0 1 Note: the value 999 means the maximum size is arbitrary. 7.5 Sample Data Records See sections 7.3.1 and 7.4.1 8. Data Organization 8.1 Data Granularity The smallest unit of model information tracked by the BOREAS Information System (BORIS) includes model code, sample input, and output files. 8.2 Data Format Model source code is written in Fortran and available as American Standard Code for Information Interchange (ASCII) text file. Sample input and output data are also stored in ASCII text files. 9. Data Manipulations 9.1 Formulae 9.1.1 Derivation Techniques and Algorithms See section 3 for key model equations. 9.2 Data Processing Sequence 9.2.1 Processing Steps First input variables for SAIL model are read in, using these inputs the SAIL subroutine is called to calculate crown reflectance and transmittance, SAIL model results are output. The geometric model inputs are read in, based on input code one of two geometric model subroutines is called. The geometric subroutine calculates scene reflectance and radiation absorptance and outputs those values. 9.2.2 Processing Changes None. 9.3 Calculations 9.3.1 Special Corrections/Adjustments When using cone-shaped crowns, if the solar zenith angle is less than the cone aspect angle, there are no shadows in the scene. This will cause the model to fail. 9.3.2 Calculated Variables Crown nadir reflectance, crown transmittance, crown hemispherical reflectance, total crown LAI, total scene LAI, scene nadir reflectance, and fraction of absorbed radiation. 9.4 Graphs and Plots None. 10. Errors 10.1 Sources of Error GeoSail is a simple model, it assumes all trees in the scene are identical and have identical reflectance and transmittance characteristics. The model also assumes that tree shadows are uniform, and are very small relative to the size of the pixel. In the model trees do not cast shadows on other trees. GeoSail accounts for multiple scattering within tree crowns, but not between crowns. 10.2 Quality Assessment 10.2.1 Model Validation by Source GeoSail has been able to model the observed patterns of red and near infrared reflectance for aspen and spruce stands over the BOREAS sites, as well as sites in the Superior National Forest in Minnesota. 10.2.2 Confidence Level/Accuracy Judgment For the open woodlands of the boreal forest, GeoSail can provide estimates of canopy reflectance within approximately 15 percent. It tends to overestimate Fapar. 10.2.3 Measurement Error for Parameters None. 10.2.4 Additional Quality Assessments None. 10.2.5 Data Verification by Data Center None. 11. Notes 11.1 Limitations of the Model GeoSail only simulates nadir views, and assumes all trees in the scene are identical. 11.2 Known Problems with the Model None. 11.3 Usage Guidance Be careful not to allow the solar zenith angle to be less that the cone aspect angle when using cone-shaped crowns. This will cause the model to fail. 11.4 Other Relevant Information None. 12. Application of the Model The GeoSail model is a useful, yet relatively simple, model that can describe canopy reflectance for forests and open woodlands. It can be used for the examination of estimation of biophysical variables from optical remote sensing data. 13. Future Modifications and Plans There may be a hyperspectral version of GeoSail in the future. 14. Software 14.1 Software Description The GeoSail model is made up of two parts: the SAIL model that calculates within crown radiative transfer, and the Geo (geometric model) part that determines amount of shadowing in a scene. The program consists of a driver that reads and writes the model input and output and calls subroutines that make up the model. This code has been compiled on Vax and Macintosh computers without problems. The SAIL model code is a modified version of the code written by Alexander (1983). The SAIL model is in subroutine sail, the sail subroutine calls a second subroutine mmult (a matrix multiplication subroutine). The geometric part of the model is based on Jasinski's geometric model (Jasinski 1990, Jasinski and Eagleson 1989 and 1990). A full description of GeoSail is in Huemmrich (1995). There are two GeoSail subroutines: geocyli and geocone, to describe tree crowns that are shaped like cylinders or cones, respectively. These subroutines use crown reflectance and transmittance values passed from the sail subroutine. 14.2 Software Access The GeoSail code is freely available, the code should be found with this documentation. 14.3 Platform Limitations The model is a fairly straightforward Fortran code, so it should be able to be compiled with most Fortran compilers. 15. Data Access 15.1 Contact for Data Center/Data Access Information These BOREAS data are available from the Earth Observing System Data and Information System (EOS-DIS) Oak Ridge National Laboratory (ORNL) Distributed Active Archive Center (DAAC). The BOREAS contact at ORNL is: ORNL DAAC User Services Oak Ridge National Laboratory (865) 241-3952 ornldaac@ornl.gov ornl@eos.nasa.gov 15.2 Procedures for Obtaining Data BOREAS data may be obtained through the ORNL DAAC World Wide Web site at http://www-eosdis.ornl.gov/ or users may place requests for data by telephone, electronic mail, or fax. 15.3 Output Products and Availability Requested data can be provided electronically on the ORNL DAAC's anonymous FTP site or on various media including, CD-ROMs, 8-MM tapes, or diskettes. The complete set of BOREAS data CD-ROMs, entitled "Collected Data of the Boreal Ecosystem-Atmosphere Study", edited by Newcomer, J., et al., NASA, 1999, are also available. 16. Output Products and Availability 16.1 Tape Products None. 16.2 Film Products None. 16.3 Other Products None. 17. References 17.1 Model Documentation None. 17.2 Journal Articles and Study Reports Alexander, L. 1983. SAIL Canopy Model Fortran Software, Lyndon B. Johnson Space Center. NASA Technical Report JSC-18899. Bégué, A. 1991. Modeling hemispherical and directional radiative fluxes in regular-clumped canopies. Remote Sens. Environ. 40(3):219-230. Goward, S.N. and K.F. Huemmrich. 1992. Vegetation canopy PAR absorptance and the normalized difference vegetation index: an assessment using the SAIL model. Remote Sensing of Environment 39:119-140. Huemmrich, K.F. 1995. An analysis of remote sensing of the fraction of absorbed photosynthetically active radiation in forest canopies. University of Maryland, Ph. D. Jasinski, M.F. 1990. Functional relation among subpixel canopy cover, ground shadow, and illuminated ground at large sampling scales. In: Society of Photo Optical Instrumentation Engineers, Orlando, FL. Jasinski, M.F. and P.S. Eagleson. 1989. The structure of red-infrared scattergrams of semivegetated landscapes. IEEE Trans. Geos. Rem. Sens. 27(4): 441-451. Jasinski, M.F. and P.S. Eagleson. 1990. Estimation of subpixel vegetation cover using red-infrared scattergrams. IEEE Trans. Geos. Rem. Sens. 28 (2): 253-267. Sellers, P. and F. Hall. 1994. Boreal Ecosystem-Atmosphere Study: Experiment Plan. Version 1994-3.0, NASA BOREAS Report (EXPLAN 94). Sellers, P. and F. Hall. 1996. Boreal Ecosystem-Atmosphere Study: Experiment Plan. Version 1996-2.0, NASA BOREAS Report (EXPLAN 96). Sellers, P., F. Hall, and K.F. Huemmrich. 1996. Boreal Ecosystem-Atmosphere Study: 1994 Operations. NASA BOREAS Report (OPS DOC 94). Sellers, P., F. Hall, and K.F. Huemmrich. 1997. Boreal Ecosystem-Atmosphere Study: 1996 Operations. NASA BOREAS Report (OPS DOC 96). Sellers, P., F. Hall, H. Margolis, B. Kelly, D. Baldocchi, G. den Hartog, J. Cihlar, M.G. Ryan, B. Goodison, P. Crill, K.J. Ranson, D. Lettenmaier, and D.E. Wickland. 1995. The boreal ecosystem-atmosphere study (BOREAS): an overview and early results from the 1994 field year. Bulletin of the American Meteorological Society. 76(9):1549-1577. Sellers, P.J., F.G. Hall, R.D. Kelly, A. Black, D. Baldocchi, J. Berry, M. Ryan, K.J. Ranson, P.M. Crill, D.P. Lettenmaier, H. Margolis, J. Cihlar, J. Newcomer, D. Fitzjarrald, P.G. Jarvis, S.T. Gower, D. Halliwell, D. Williams, B. Goodison, D.E. Wickland, and F.E. Guertin. 1997. BOREAS in 1997: Experiment overview, scientific results, and future directions. Journal of Geophysical Research 102(D24):28,731- 28,769. Verhoef, W. 1984. Light scattering by leaf layers with application to canopy reflectance modeling: the SAIL model. Remote Sens. Environ. 16:125-141. 17.3 Archive/DBMS Usage Documentation 18. Glossary of Terms None. 19. List of Acronyms ASCII - American Standard Code for Information Interchange BOREAS - BOReal Ecosystem-Atmosphere Study BORIS - BOREAS Information System DAAC - Distributed Active Archive Center EOS - Earth Observing System EOSDIS - EOS Data and Information System Fapar - Fraction of Absorbed Photosynthetically Active Radiation GSFC - Goddard Space Flight Center LAI - Leaf Area Index NASA - National Aeronautics and Space Administration ORNL - Oak Ridge National Laboratory PAR - Photosynthetically Active Radiation RSS - Remote Sensing Science SAIL - Scattering from Arbitrarily Inclined Leaves TE - Terrestrial Ecology URL - Uniform Resource Locator 20. Document Information 20.1 Document Revision Date Written: 15-Nov-1999 Revised: 29-Nov-1999 20.2 Document Review Date(s) BORIS Review: 29-Nov-1999 Science Review: 20.3 Document ID 20.4 Citation When using the model, please include the following acknowledgment: Huemmrich, K.F. 1995. An analysis of remote sensing of the fraction of absorbed photosynthetically active radiation in forest canopies. University of Maryland, Ph. D. 20.5 Document Curator 20.6 Document URL Keywords CANOPY REFLECTANCE MODEL GEOMETRIC MODEL TURBID MEDIA MODEL REFLECTANCE TE18_GeoSail.doc 11/22/99