Skip to content

Latest commit

 

History

History
862 lines (689 loc) · 39.2 KB

SA_ch04_raster.md

File metadata and controls

862 lines (689 loc) · 39.2 KB

<< Previous Chapter - Home - Next Chapter >>

Chapter 4. Raster Tools

Contents

Introduction

The Spatial Allocator (SA) Raster Tools system is designed to process image or raster spatial data sets in SA. It contains programs to process various kinds of spatial data for meteorological and air quality modeling, particularly within the Weather Research and Forecasting (WRF) (http://www2.mmm.ucar.edu/wrf/users/) and Community Multiscale Air Quality (CMAQ) (http://www.cmascenter.org/cmaq/) modeling systems. The Raster Tools include land cover data processing tools, satellite cloud and aerosol product processing tools, agricultural fertilizer modeling tools, a domain grid shapefile generation tool, and other utilities.

All sample script files for the SA Raster Tools are stored in the raster_scripts directory of the installed Spatial Allocator system.

Compiling and Installation

Users who have difficulties running the tools with the compiled libraries contained within the downloaded Spatial Allocator system should do the following:

  • delete installed open-source library directories under the ./src/libs directory
  • download new source packages and install them under the ./libs directory
  • compile downloaded packages and install them under {package_path}/local, following the src/libs/README file
  • modify paths in ./bin/sa_setup.csh and ./src/raster/Makefile
  • in ./src/raster, do the following:
make clean
make
make install

Defining Domains

The SA Raster Tools define the modeling domain using the following environment variables:

  • GRID_PROJ – defines the domain grid projection using the PROJ4 projection description format, for a full list see: (http://spatialreference.org/) or on the PROJ4 wiki: (http://proj4.org/index.html). The following sample projection descriptions are used to match the projections in WRF:

    • Lambert Conformal Conic: +proj=lcc +a=6370000.0 +b=6370000.0 +lat_1=33 +lat_2=45 +lat_0=40 +lon_0=-97
    • Polar stereographic: +proj=stere +a=6370000.0 +b=6370000.0 +lat_ts=33 +lat_0=90 +lon_0=-97 +k_0=1.0
    • Mercator: +proj=merc +a=6370000.0 +b=6370000.0 +lat_ts=33 +lon_0=0
    • Geographic: +proj=latlong +a=6370000.0 +b=6370000.0

  • GRID_ROWS – number of rows of grid cells in the domain

  • GRID_COLUMNS – number of columns of grid cells in the domain

  • GRID_XCELLSIZE – grid cell size in x direction

  • GRID_YCELLSIZE – grid cell size in y direction

  • GRID_XMIN – minimum x of the domain (lower left corner of the domain)

  • GRID_YMIN – minimum y of the domain (lower left corner of the domain)

  • GRID_NAME – name of the domain, which is required by some of the tools

For WRF simulations, GRID_XMIN and GRID_YMIN can be computed using the first point longitude and latitude from the global attributes corner_lons and corner_lats in the domain’s WRF GEOGRID output file. For instance, to compute a WRF Lambert Conformal Conic (LCC) domain with the GEOGRID output file attributes

 :corner_lats = 20.85681f, 52.1644f, 50.63151f, 19.88695f, 20.84302f...
 :corner_lons = -121.4918f, -135.7477f, -53.21942f, -69.02478f, -121.5451f…

use the cs2cs utility in the PROJ4 library directly at the command line (after installing the SA system):

cs2cs +proj=latlong +a=6370000.0 +b=6370000.0 +to +proj=lcc +a=6370000.0 +b=6370000.0 +lat_1=33 +lat_2=45 +lat_0=40 +lon_0=-97

Then input the following long/lat as input

-121.4918 20.85681

the cs2c program will output the following x/y value as output:

-2622003.85 -1793999.28 0.00

Minimum x and y for the domain would be computed as follows:

GRID_XMIN = -2622003.85 - GRID_XCELLSIZE / 2
GRID_YMIN = -1793999.28 - GRID_YCELLSIZE / 2

Land Cover Data Processing Tools

There are three land cover processing tools in the SA Raster Tools:

All of the example scripts listed in this section are in the SA_HOME/raster_scripts directory.

1. NLCD and MODIS Land Cover Generation

The computeGridLandUse.exe tool is used to generate land cover data for the upgraded WRF/CMAQ Pleim-Xiu Land Surface Model (PX LSM) in the current WRF model release, by directly using downloaded 2001, 2006, or 2011 National Land Cover Data (NLCD) land cover data, the NASA Moderate Resolution Imaging Spectroradiometer (MODIS) land cover products MCD12Q1 or MOD12Q1, and NASA MODIS LAI/FPAR products (e.g. MCD15A2H, MOD15A2, MOD15A2GFS). This tool generates 40 land cover classes (20 from MODIS and 20 from NLCD), and MODIS LAI/FPAR data for each land cover type and whole grid area.

This tool requires the following data sets:

  • NLCD land cover, canopy and imperviousness data:

  • MODIS tiled land cover data: MCD12Q1

    • SA Scripts: NLCD_MODIS_processor.csh and landuseTool_WRFCMAQ_BELD4.csh

    • URL: https://ladsweb.modaps.eosdis.nasa.gov/search/

    • Instructions: Select the following from the download page:

    • PRODUCTS:

      • Combined Terra & Aqua MODIS
      • Combined Land Level 3/Level4 yearly Tiled Products (keyword: MCD12Q1) - MCD12Q1 - MODIS/Terra+Aqua Land Cover Type Yearly L3 Global 500m SIN Grid

    • TIME

      • Temporal Type: Date and Time Range
      • Set time period for downloading (typically 1 year, example Jan 1, 2011 - Dec. 31, 2011)

    • LOCATION (Choose either)

      • Coordinate System: Latitude/Longitude
        • Type in extents in degree or use a predefined region
      • or
      • Country - (example United States)

    • FILES:

      • Select All

    • REVIEW and ORDER:

      • Select Delivery Method
        • Submit Order

  • List of land cover data sets to be processed – this file has to have fixed header formats. Provided in the data directory are sample files for CMAQ 12-km domain 2001, 2006 and 2011 modeling:

nlcd_modis_files_2001.txt
nlcd_modis_files_2006.txt
nlcd_modis_files_2011.txt.

Users have to modify the list file based on their NLCD and MODIS data location and names.

To run the computeGridLandUse tool, users can use the following script file, which has all of the required environment variables:

NLCD_MODIS_processor.csh

The tool generates one ASCII file and one NetCDF file:

  • The ASCII file contains the imperviousness, canopy, and land cover percent variables (if the user set all land cover data to “YES” when running the script file) for each grid cell, in comma-separated-values (CSV) format.
  • The NetCDF file contains imperviousness, canopy, and land cover fraction variables plus land/water mask and other variables that are similar to those in the WRF GEOGRID land cover output files. The land cover percentage variable contains the 40 classes in Table 1.

Table 1. NLCD/MODIS output land cover classes from the computeGridLandUse tool.

Array Index MODIS Class IGBP (Type 1) Class Name Array Index NLCD Class Class Name
1 1 Evergreen Needleleaf forest 21 11 Open Water
2 2 Evergreen Broadleaf forest 22 12 Perennial Ice/Snow
3 3 Deciduous Needleleaf forest 23 21 Developed - Open Space
4 4 Deciduous Broadleaf forest 24 22 Developed - Low Intensity
5 5 Mixed forest 25 23 Developed - Medium Intensity
6 6 Closed shrublands 26 24 Developed High Intensity
7 7 Open shrublands 27 31 Barren Land (Rock/Sand/Clay)
8 8 Woody savannas 28 41 Deciduous Forest
9 9 Savannas 29 42 Evergreen Forest
10 10 Grasslands 30 43 Mixed Forest
11 11 Permanent wetlands 31 51 Dwarf Scrub
12 12 Croplands 32 52 Shrub/Scrub
13 13 Urban and built-up 33 71 Grassland/Herbaceous
14 14 Cropland/Natural vegetation mosaic 34 72 Sedge/Herbaceous
15 15 Snow and ice 35 73 Lichens
16 16 Barren or sparsely vegetated 36 74 Moss
17 0 Water 37 81 Pasture/Hay
18 18 Reserved (e.g., Unclassified) 38 82 Cultivated Crops
19 19 Reserved (e.g., Fill Value ) 39 90 Woody Wetlands
20 20 Reserved 40 95 Emergent Herbaceous Wetlands

2. NLCD and MODIS Land Cover and MODIS LAI Generation

The computeGridLandUse_LAI_MODIS.exe tool is used to generate land cover data for the WRF/CMAQ Pleim-Xiu Land Surface Model (PX LSM) in the current WRF model release, by directly using downloaded 2001, 2006, or 2011 National Land Cover Data (NLCD) land cover data, the NASA Moderate Resolution Imaging Spectroradiometer (MODIS) land cover products MCD12Q1 or MOD12Q1, and NASA MODIS LAI/FPAR products (e.g. MCD15A2H, MOD15A2, MOD15A2GFS). This tool generates 40 land cover classes (20 from MODIS and 20 from NLCD) and MODIS LAI/FPAR data for each land cover type and whole grid area.

This tool requires the following data sets:

  • NLCD land cover, canopy, and imperviousness data – can be obtained from https://www.mrlc.gov/.

  • MODIS land cover data sets – can be obtained through https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12q1. The tool can process MCD12Q1 data at 500 m from Combined Terra and Aqua MODIS, or can process MOD12Q1 data at 1 km from Terra MODIS.

  • MODIS LAI/FPAR products – can be obtained through https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mod15a2h_v006.The tool can process any of MCD15A2H, MOD15A2, and MOD15A2GFS MODIS vegetation products.

    List of land cover data sets to be processed – this file has to have fixed header formats. Provided in the data directory are sample files for CMAQ 12-km domain 2001, 2006 and 2011 modeling: nlcd_modis_files_2001.txt, nlcd_modis_files_2006.txt, and nlcd_modis_files_2011.txt. Users have to modify the list file based on their NLCD and MODIS data location and names.

To run the computeGridLandUse_LAI_MODIS tool, users can use the following sample script file, which has all of the required environment variables:

MODIS_landcover_LAI_data_processor.csh

The tool generates one ASCII file and one NetCDF file:

  • The ASCII file contains the imperviousness, canopy, and land cover percent variables (if the user set all land cover data to “YES” when running the script file) for each grid cell, in comma-separated-values (CSV) format.

The NetCDF file contains imperviousness, canopy, and land cover fraction variables plus land/water mask and other variables that are similar to those in the WRF GEOGRID land cover output files. The land cover percentage variable contains the 40 classes as listed in Table 1. In addition, MODIS LAI and FPAR variables for each landcover type and average at each grid cell are included in the NetCDF file.

3. BELD4 Land Cover Generation

The BELD4 data with land cover, tree, and crop percentages can be computed using the computeGridLandUse_beld4.exe tool with directly downloaded USGS NLCD data sets, NASA MODIS land cover (MCD12Q1 or MOD12Q1) data tiles and tree and crop fractions at the county level. The following sample script file contains all of the required environment variables for running the tool:

landuseTool_WRFCMAQ_BELD4.csh.

This tool requires the following data sets:

  • Downloaded USGS NLCD data sets, including land cover, imperviousness, and canopy, can be obtained from the NLCD web site: http://www.mrlc.gov/nlcd2006.php.
  • MODIS land cover tiles (MCD12Q1 or MOD12Q1) – can be obtained from the NASA MODIS land products web site: http://modis-land.gsfc.nasa.gov/landcover.html.
  • List of land cover data sets to be processed – this file has to be fixed format with the data set headers included. Provided in the data directory are sample files for CMAQ 12-km domain 2001, 2006 and 2011 modeling: nlcd_modis_files_2001.txt, nlcd_modis_files_2006.txt, and nlcd_modis_files_2011.txt. Users have to modify the list file based on their NLCD and MODIS data location and names.
  • BELD3 FIA tree fraction table at county level – provided in the data directory: beld3-fia.dat.
  • National Agricultural Statistics Service (NASS) crop fraction tables at county level – provided in the data directory: nass2001_beld4_ag.dat for the 2001 NASS; nass2006_beld4_ag.dat for the 2006 NASS.
  • Canada crop fraction table at Census-division level – provided in the data directory: can01_beld4_ag.dat for the 2001 Census of Agriculture; can06_beld4_ag.dat for the 2006 Census of Agriculture.
  • List of land cover, tree, and crop classes for the BELD4 tool – provided in the data direc­tory: beld4_class_names_40classes.txt.
  • U.S. county shapefile – provided in the data directory: county_pophu02_48st.shp.
  • Canada Census-division shapefiles – provided in the data directory: can2001_cd_sel.shp for the 2001 Census; can2006_cd_sel.shp for the 2006 Census.

The tool generates one ASCII file and one NetCDF file:

  • The ASCII file contains the imperviousness, canopy, and land cover fraction variables (if the user set all land cover data to “YES” when running the script file) for each grid cell, in CSV format.
  • The NetCDF file contains imperviousness, canopy, land cover, tree, and crop percentage variables as well as land/water mask and other variables that are similar to those in the WRF GEOGRID land cover output files.

The land cover data generated by applying this tool are used in CMAQ bidirectional ammonia flux modeling and are used in CMAQ biogenic, land surface, and dry deposition modeling. The land cover percentage array in the output contains 20 NLCD land cover classes and 20 MODIS IGBP land cover classes (see Table 1). The tree percentage variable in the NetCDF output file contains the 194 BELD4 tree classes shown in Table 2, and the crop percentage variable contains the 42 crops listed in Table 3.

Table 2. BELD4 tree classes.

Index Variable Index Variable Index Variable Index Variable
1 Acacia 40 Hackberry 79 Oak_bur 118 Paulownia
2 Ailanthus 41 Hawthorn 80 Oak_CA_black 119 Pawpaw
3 Alder 42 Hemlock 81 Oak_CA_live 120 Persimmon
4 Apple 43 Hickory 82 Oak_CA_white 121 Pine_Apache
5 Ash 44 Holly_American 83 Oak_canyon_live 122 Pine_Austrian
6 Basswood 45 Hornbeam 84 Oak_chestnut 123 Pine_AZ
7 Beech 46 Incense_cedar 85 Oak_chinkapin 124 Pine_Bishop
8 Birch 47 Juniper 86 Oak_delta_post 125 Pine_blackjack
9 Bumelia_gum 48 KY_coffeetree 87 Oak_Durand 126 Pine_brstlcone
10 Cajeput 49 Larch 88 Oak_Emery 127 Pine_chihuahua
11 Califor-laurel 50 Loblolly_bay 89 Oak_Engelmann 128 Pine_Coulter
12 Cascara-buckthor 51 Madrone 90 Oak_evergreen_sp 129 Pine_digger
13 Castanea 52 Magnolia 91 Oak_Gambel 130 Pine_Ewhite
14 Catalpa 53 Mahogany 92 Oak_interio_live 131 Pine_foxtail
15 Cedar_chamaecyp 54 Maple_bigleaf 93 Oak_laurel 132 Pine_jack
16 Cedar_thuja 55 Maple_bigtooth 94 Oak_live 133 Pine_Jeffrey
17 Chestnut_buckeye 56 Maple_black 95 Oak_Mexicanblue 134 Pine_knobcone
18 Chinaberry 57 Maple_boxelder 96 Oak_Northrn_pin 135 Pine_limber
19 Cypress_cupress 58 Maple_FL 97 Oak_Northrn_red 136 Pine_loblolly
20 Cypress_taxodium 59 Maple_mtn 98 Oak_nuttall 137 Pine_lodgepole
21 Dogwood 60 Maple_Norway 99 Oak_OR_white 138 Pine_longleaf
22 Douglas_fir 61 Maple_red 100 Oak_overcup 139 Pine_Monterey
23 East_hophornbean 62 Maple_RkyMtn 101 Oak_pin 140 Pine_pinyon
24 Elder 63 Maple_silver 102 Oak_post 141 Pine_pinyon_brdr
25 Elm 64 Maple_spp 103 Oak_scarlet 142 Pine_pinyon_cmn
26 Eucalyptus 65 Maple_striped 104 Oak_scrub 143 Pine_pitch
27 Fir_balsam 66 Maple_sugar 105 Oak_shingle 144 Pine_pond
28 Fir_CA_red 67 Mesquite 106 Oak_Shumrd_red 145 Pine_ponderosa
29 Fir_corkbark 68 Misc-hardwoods 107 Oak_silverleaf 146 Pine_red
30 Fir_fraser 69 Mixed_conifer_sp 108 Oak_Southrn_red 147 Pine_sand
31 Fir_grand 70 Mountain_ash 109 Oak_spp 148 Pine_scotch
32 Fir_noble 71 Mulberry 110 Oak_swamp_cnut 149 Pine_shortleaf
33 Fir_Pacf_silver 72 Nyssa 111 Oak_swamp_red 150 Pine_slash
34 Fir_SantaLucia 73 Oak_AZ_white 112 Oak_swamp_white 151 Pine_spruce
35 Fir_Shasta_red 74 Oak_bear 113 Oak_turkey 152 Pine_sugar
36 Fir_spp 75 Oak_black 114 Oak_water 153 Pine_Swwhite
37 Fir_subalpine 76 Oak_blackjack 115 Oak_white 154 Pine_tablemtn
38 Fir_white 77 Oak_blue 116 Oak_willow 155 Pine_VA
39 Gleditsia_locust 78 Oak_bluejack 117 Osage-orange 156 Pine_Washoe

Table 3. BELD4 crop classes.

Index Variable Index Variable Index Variable
1 Hay 15 Cotton 29 SorghumSilage
2 Hay_ir 16 Cotton_ir 30 SorghumSilage_ir
3 Alfalfa 17 Oats 31 Soybeans
4 Alfalfa_ir 18 Oats_ir 32 Soybeans_ir
5 Other_Grass 19 Peanuts 33 Wheat_Spring
6 Other_Grass_ir 20 Peanuts_ir 34 Wheat_Spring_ir
7 Barley 21 Potatoes 35 Wheat_Winter
8 Barley_ir 22 Potatoes_ir 36 Wheat_Winter_ir
9 BeansEdible 23 Rice 37 Other_Crop
10 BeansEdible_ir 24 Rice_ir 38 Other_Crop_ir
11 CornGrain 25 Rye 39 Canola
12 CornGrain_ir 26 Rye_ir 40 Canola_ir
13 CornSilage 27 SorghumGrain 41 Beans
14 CornSilage_ir 28 SorghumGrain_ir 42 Beans_ir

4. Current and Future Development for the Land Cover Data Processing Tools

  • Enhance the tool to use the released NLCD 2011 data sets with created 2011 crop tables for both US and Canada.
  • Use USDA’s NLCD Cropland Data Layer (CDL) data instead of NASS crop fractions at the county level for the BELD4 data tool. This will support the use of USDA crop spatial coverage NLCD data instead of county-based crop census data in computing crop fractions within each grid cell.

Satellite Cloud and Aerosol Product Processing Tools

1. GOES Cloud Product Processing Tool

The GOES data tool processes the Geostationary Operational Environmental Satellite (GOES) data downloaded from the Earth System Science Center (ESSC) at the University of Alabama in Huntsville. The GEOS data web site is http://satdas.nsstc.nasa.gov

Downloaded GOES data need to be stored under subdirectories named using this format: gp_YYYYMMDD. The ./util/goes_untar.pl utility can be used to unzip downloaded GOES data (daily tar files) into the daily directories required by the tool.

The following sample script file contains all of the required environment variables for running the tool:

allocateGOES2WRFGrids.csh.

The tool contains the following three programs:

  • correctGOESHeader.exe – to correct GOES data position shifting by redefining a new Earth radius and new image extent. The program converts GOES data in Grib (i.e., .grb) format to files in ERDAS Imagine (i.e., .img) format with corrections.
  • computeGridGOES.exe – to regrid corrected Imagine-format GOES data to a defined grid domain.
  • toDataAssimilationFMT.exe – to convert the gridded NetCDF file into a format suitable for WRF assimilation.

The released GOES data has changed to ASCII format from GRIB format last year. We plan to update the tool in the coming months.

When running the GOES cloud product processing tool, the Geospatial Data Abstraction Library (GDAL) will generate the following messages:

Warning: Inside GRIB2Inventory, Message # 2
ERROR: Ran out of file reading SECT0

These messages do not indicate any errors in regridding and so can be ignored.

2. MODIS Level 2 Cloud/Aerosol Products Tool

The MODIS Level 2 (swath) cloud and aerosol products tool processes MODIS L2 cloud or aerosol products for a defined grid domain. MODIS data in HDF4 format can be downloaded from the NASA Level 1 and Atmosphere Archive and Distribution System (LAADS) web site: http://ladsweb.nascom.nasa.gov/data/search.html.

MODIS cloud product variables contain 5-km and 1-km data. To use this regridding tool, users need to download the following cloud data and Level 1 Geolocation 1-km data into the input directory:

  • MOD06_L2 and MOD03 (Level 1 Geolocation 1-km ) for Terra, or
  • MYD06_L2 and MYD03 (Level 1 Geolocation 1-km ) for Aqua

The following download options can be selected during the download process:

MODIS Cloud:

  • Select Level 2 products and select L2 Cloud products
  • Select time: "your download time period"
  • Collection 5
  • Select Latitude/Longitude with the above geographic extent
  • Coverage options: select day, night, and both (all)
  • Select all other defaults and click search
  • Display all files
  • Download files into one directory

MODIS Geolocation 1-km:

  • Select Level 1 products and select O3 Geolocation - 1km
  • Select time: "same as cloud products"
  • Collection 5
  • Select Latitude/Longitude with the above geographic extent
  • Coverage options: select day, night,and both (all)
  • Display all files
  • Download all files into the MODIS Cloud file directory

MODIS aerosol products contain variable data at 10-km resolution (nadir). Users need to download MOD04 for Terra or MYD04 for Aqua into the input data directory. The download options below can be selected when downloading Terra aerosol products. Downloading Aqua aerosol products involves similar options. The tool generates one NetCDF file for the defined domain.

Aerosol Product Data Access:

  • Select Terra MODIS
  • MODIS Aerosol products
  • Select Level 2 products and select L2 aerosol poduct
  • Select time: "your download time period"
  • Collection: 5
  • Select Latitude/Longitude with area longitude and latitude extent
  • Coverage options: select day, night, and both (all)
  • Select all other defaults and click search
  • Display all files
  • Download all files into one directory

Users can modify the following sample script file provided for regridding the MODIS cloud data:

allocateMODISL2CloudVars2Grids.csh

3. OMI Level 2 Product Tool

The OMI Level 2 product (swath) tool is used to regrid Ozone Monitoring Instrument (OMI) L2 aerosol and NO2 products for a defined grid domain. The input data can be downloaded from the NASA mirador site:

http://mirador.gsfc.nasa.gov/cgi-bin/mirador/presentNavigation.pl?tree= project&project=OMI.

The downloaded data are in HDF5 format and should be stored in one directory, which is defined in the following sample script file:

allocateOMIL2vars2Grids.csh

4. OMI L2G and L3 Product Tools

The OMI L2G and L3 product tools process the following OMI products:

  • OMI L3 aerosol products (OMAEROe) in HDF4
  • OMI NO2 L2G products (OMNO2G) in HDF4
  • OMI NO2 L3 products (NO2TropCS30) in HDF5

The data can be downloaded from the NASA Giovanni web site: http://gdata1.sci.gsfc.nasa.gov/daac-bin/G3/gui.cgi?instance_id=omi

OMI product information can be viewed from http://disc.sci.gsfc.nasa.gov/giovanni/additional/ users-manual/G3_manual_Chapter_10_OMIL2G.shtml#what_l2g and from ftp://aurapar2u.ecs .nasa.gov/data/s4pa//Aura_OMI_Level2/OMAERUV.003/doc/README.OMI_DUG.pdf

The following sample script can be modified for regridding:

allocateOMIvar2Grids.csh

Agricultural Fertilizer Modeling Tools

There are four tools that can be used when performing Environmental Policy Integrated Climate (EPIC) modeling; they generate gridded agricultural fertilizer data to be used in CMAQ bidirectional NH3 flux modeling. These tools are the EPIC site information generation tool, the MCIP/CMAQ-to-EPIC tool, the EPIC-to-CMAQ tool, and the EPIC yearly extraction tool](#epic_yearly). They can be called from the Fertilizer Emission Scenario Tool for CMAQ (FEST-C) interface (http://www.cmascenter.org/fest-c/) based on user input information, and can be run by script files with defined environment variables at the command line.

1. EPIC Site Information Generation Tool

This tool generates three CSV data files that are needed to create EPIC site databases for a user-defined domain:

  • EPICSites_Info.csv – contains GRIDID, XLONG, YLAT, ELEVATION, SLOPE_P, HUC8, REG10, STFIPS, CNTYFIPS, GRASS, CROPS, CROP_P, COUNTRY, and COUNTRY-PROVINCE items.
  • EPICSites_Crop.csv – contains GRIDID, 42 crop acreages, COUNTRY, and HUC8 items.
  • allSites_Crop.csv - contains GRIDID, 42 crop acreages, COUNTRY, and HUC8 items for all grid cells.

The tool processes the set of input spatial data files below, which have been modified specifically for use with the tool and can be obtained from the CMAS:

  • BELD4 file for the domain (beld4_cmaq12km_2006.nc)
  • U.S. county shapefiles (co99_d00_conus_cmaq_epic.shp)
  • North American State political boundary shapefile (na_bnd_camq_epic.shp)
  • U.S. 8-digit HUC shapefile (conus_hucs_8_cmaq.shp)
  • Elevation image file (na_dem_epic.img)
  • Slope image file (na_slope_epic.img)

Users can follow the sample script file below, which has all of the environment variables required for running the tool from the command line:

generateEPICSiteData.csh

2. MCIP/CMAQ-to-EPIC Tool

This tool generates EPIC daily weather and nitrogen deposition data files from MCIP meteorology and CMAQ nitrogen deposition files for EPIC modeling sites. The input MCIP and CMAQ data are stored in two directories defined by the environment variables DATA_DIR and DATA_DIR_CMAQ.

MCIP output files must have names of the format METCRO2D*{date} (e.g., METCRO2D_020725). The date format can be in one of the following formats:

YYYYMMDD *or* YYMMDD *or* YYYYDDD *or* YYDDD

CMAQ dry and wet deposition files must have names of the format *DRYDEP*{date} and *WETDEP*{date} (e.g., CCTM_N4a_06emisv2soa_12km_wrf.DRYDEP.20020630 and CCTM_N4a_06emisv2soa_12km_wrf.WETDEP1.20020630). The date can be in any of the formats listed above.

Deposition inputs for EPIC modeling can take one of the following three inputs:

  • Directory containing a CMAQ dry and wet deposition file
  • Zero – assume zero nitrogen deposition
  • Default – assume nitrogen mix ratio of 0.8 ppm for wet default deposition computation

The input site location file defined by the environment variable EPIC_SITE_FILE has to be a CSV file, with the first three items being site ID, longitude, and latitude.

allSites_Crop.csv generated by the EPIC Site Information Generation Tool described above is used as EPIC_SITE_FILE for FEST-C applications.

The tool generates three outputs:

  • dailyWETH directory containing EPIC daily weather and nitrogen deposition files with names of the format “grid ID”.dly (e.g., 96.dly). The daily file contains the 14 variables listed in Table 4.
  • NetCDF file with daily weather and nitrogen deposition data for all grid cells.
  • EPICW2YR.2YR, to be used for daily weather file input list in EPIC modeling.

Table 4. EPIC daily weather and nitrogen deposition variables.

Index Variable Index Variable
1 Year 8 Daily Average Relative Humidity
2 Month 9 Daily Average 10m Windspeed (m s^-1)
3 Day 10 Daily Total Wet Oxidized N (g/ha)
4 Daily Total Radiation (MJ m^02) 11 Daily Total Wet Reduced N (g/ha)
5 Daily Maximum 2m Temperature (C) 12 Daily Total Dry Oxidized N (g/ha)
6 Daily minimum 2m temperature (C) 13 Daily Total Dry Reduced N (g/ha)
7 Daily Total Precipitation (mm) 14 Daily Total Wet Organic N (g/ha)

Users can follow the sample script file below, which has all of the environment variables required for running the tool from the command line window:

generateEPICsiteDailyWeatherNdep.csh

The following are two versions of the tool computeSiteDailyWeather.cpp

1. computeSiteDailyWeather.cpp_beforecmaq52

2. computeSiteDailyWeather.cpp_cmaq52

The default version is linked to the version: computeSiteDailyWeatehr.cpp_cmaq52.

Users should change the link to the computeSiteDailyWeather.cpp_beforecmaq52 if a version of CMAQ prior to CMAQv5.2 was used to generate the N deposition input files.

3. EPIC-to-CMAQ Tool

This tool processes merged daily output from EPIC simulations for the 42 crops defined for the BELD4 tool output. It generates two types of outputs in NetCDF format for CMAQ bidirectional NH3 modeling:

  • soil output file
  • EPIC daily output files

The 13 variables contained in the soil output file are listed in Table 5.

Table 5. EPIC-to-CMAQ soil output variables.

Index Name Soil Variable Index Name Soil Variable
1 L1_SoilNum Soil Number (none) 8 L2_Bulk_D Layer2 Bulk Density (t/m**3)
2 L1_Bulk_D Layer1 Bulk Density (t/m**3) 9 L2_Wilt_P Layer2 Wilting Point (m/m)
3 L1_Wilt_P Layer1 Wilting Point(m/m) 10 L2_Field_C Layer2 Field Capacity (m/m)
4 L1_Field_C Layer1 Field Capacity (m/m) 11 L2_Porocity Layer2 Porocity (%)
5 L1_Porocity Layer1 Porocity (%) 12 L2_PH Layer2 PH (none)
6 L1_PH Layer1 PH (none) 13 L2_Cation Layer2 Cation Ex (cmol/kg)
7 L1_Cation Layer1 Cation Ex (cmol/kg)

EPIC daily output files for CMAQ contain the 41 variables listed in Table 6.

The following sample script file with all of the required environment variables can be modified and run at the command line:

epic2CMAQ.csh

Table 6. EPIC for CMAQ daily output variables.

Index Name Variable Index Name Variable
1 DN N-NO3 Denitrification (kg/ha) 22 L2_NH3 Layer2 Ammonia (kg/ha)
2 DN2* N-N2O from NO3 Denitrification (kg/ha) 23 L2_ON Layer2 Organic N (kg/ha)
3 HMN OC Change by Soil Respiration (kg/ha) 24 L2_C Layer2 Carbon (kg/ha)
4 NFIX N Fixation (kg/ha) 25 L2_NITR Layer2 N - Nitrified NH3 (kg/ha)
5 GMN N Mineralized (kg/ha) 26 T1_DEP Layer Depth (m)
6 YW Wind Erosion (ton/ha) 27 T1_BD Layer Bulk Density (t/m**3)
7 FPO Organic P Fertilizer (kg/ha) 28 T1_NO3 Layer N - Nitrate (kg/ha)
8 FPL Labile P Fertilizer (kg/ha) 29 T1_NH3 Layer N - Ammonia (kg/ha)
9 MNP P Mineralized (kg/ha) 30 T1_ON Layer Organic N (kg/ha)
10 L1_DEP Layer1 Depth (m) 31 T1_C Layer Mineral C (kg/ha)
18 L1_BD Layer1 Bulk Density (t/m**3) 32 T1_NITR Layer N - Nitrified NH3 (kg/ha)
12 L1_SW Layer1 Soil Moisture (mm) 33 T1_ANO3 Layer1 N-NH3 AppRate (kg/ha)
13 L1_NO3 Layer1 N - Nitrate (kg/ha) 34 T1_ANH3 Layer1 N-NH3 AppRate (kg/ha)
14 L1_NH3 Layer1 N - Ammonia (kg/ha) 35 T1_AON Layer1 ON AppRate (kg/ha)
15 L1_ON Layer1 Organic N (kg/ha) 36 L2_ANH3 Layer2 N-NO3 AppRate (kg/ha)
16 L1_C Layer1 Carbon (kg/ha) 37 L2_ANH3 Layer2 N-NH3 AppRate (kg/ha)
17 L1_NITR Layer1 N - Nitrified NH3 (kg/ha) 38 L2_AON Layer2 ON AppRate (kg/ha)
18 L2_DEP Layer1 Depth (m) 39 LAI Leaf Area Index (none)
19 L2_BD Layer2 Bulk Density (t/m**3) 40 CPHT Crop Height(m)
20 L2_SW Layer2 Soil Moisture (mm) 41 FBARE Bare Land Fraction for Wind Erosion (Fraction)
21 L2_NO3 Layer2 N - Nitrate (kg/ha)

Note: EPIC is a daily timestep model while the CMAQ bidirectional NH3 flux model is at a time scale which could be less than 10 minutes.

4. EPIC Yearly Extraction Tool

This tool is used primarily to provide data for performing quality assurance (QA) for EPIC runs.

  • For EPIC spin-up runs, it extracts average EPIC values from the last five years of the spin-up simulations.
  • For EPIC application runs, it extracts application-year EPIC variables.

In both cases, the tool outputs one crop-specific NetCDF file with 48 variables and one crop-weighted NetCDF file with 39 variables; Table 7 shows the two lists of variables.

Table 7. EPIC yearly extraction output variables.

epic2cmaqyear.nc - crop specific output

Index Name Variable Index Name Variable
1 GMN N Mineralized (kg/ha) 25 FTP P Applied (kg/ha)
2 NMN Humus Mineralization (kg/ha) 26 IRGA* Irrigation Volume Applied (mm)
3 NFIX N Fixation (kg/ha) 27 WS Water Stress Days (days)
4 NITR N - Nitrified NH3 (kg/ha) 28 NS N Stress Days (days)
5 AVOL N - Volatilization (kg/ha) 29 IPLD Planting Date (Julian Date)
6 DN N-NO3 Denitrification (kg/ha) 30 IGMD Germination Date (Julian Date)
7 YON N Loss with Sediment (kg/ha) 31 IHVD Harvest Date (Julian Date)
8 QNO3 N Loss in Surface Runoff (kg/ha) 32 YP P Loss with Sediment (kg/ha)
9 SSFN N in Subsurface Flow (kg/ha) 33 QAP Labile P Loss in Runoff (kg/ha)
10 PRKN N Loss in Percolate (kg/ha) 34 YW Wind Erosion (ton/ha)
11 FNO N - Organic Fertilizer (kg/ha) 35 Q* Runoff (mm)
12 FNO3 N - Nitrate Fertilize (kg/ha) 36 SSF Subsurface flow (mm)
13 FNH3 N - Ammonia Fertilize (kg/ha) 37 PRK Percolation (mm)
14 OCPD Organic Carbon in Plow Layer (mt/ha) 38 PRCP Rainfall (mm)
15 TOC Organic Carbon in Soil Profile (mt/ha) 39 PET Potential Evapotranspiration (mm)
16 TNO3 Total NO3 in Soil Profile (kg/ha) 40 ET Evapotranspiration (mm)
17 DN2 N-N2O from NO3 Denitrification (kg/ha) 41 QDRN Drain Tile Flow (mm)
18 YLDG Grain Yield (t/ha) 42 MUSL Water erosion (ton/ha)
19 T_YLDG T-Grain Yield (1000ton) 43 DRNN Nitrogen in drain tile flow (kg/ha)
20 YLDF Forage Yield (t/ha) 44 DRNP P in Drain Tile Flow (kg/ha)
21 T_YLDF T-Forage Yield (1000ton) 45 PRKP P in Percolation (kg/ha)
22 YLN N Used by Crop (kg/ha) 46 FPO Organic P Fertilizer (kg/ha)
23 YLP P Used by Crop (kg/ha) 47 FPL Labile P Fertilizer (kg/ha)
24 FTN N Applied (kg/ha) 48 MPN P Mineralized (kg/ha)

epic2cmaq_year_total.nc - crop weighted output

Index Name Variable Index Name Variable
1 T_GMN N Mineralized (mt - metric ton) 21 T_FTP P Applied (mt)
2 T_NMN Humus Mineralization (mt) 22 T_IRGA* Irrigation Volume Applied (mm)
3 T_NFIX N Fixation (mt) 23 T_YP T -P Loss with Sediment (mt)
4 T_NITR N - Nitrified NH3 (mt) 24 T_QAP T -Labile P Loss in Runoff (mt)
5 T_AVOL N - Volatilization (mt) 25 T_YW T -Wind Erosion (1000ton)
6 T_DN N-NO3 Denitrification (mt) 26 T_Q* T -Runoff (mm)
7 T_YON N Loss with Sediment (mt) 27 T_SSF T - Subsurface flow (mm)
8 T_QNO3 N Loss in Surface Runoff (mt) 28 T_PRK T - Percolation (mm)
9 T_SSFN N in Subsurface Flow (mt) 29 T_PRCP T-Rainfall (mm)
10 T_PRKN N Loss in Percolate (mt) 30 T_PET T - Potential Evapotranspiration (mm)
11 T_FNO N - Organic Fertilizer (mt) 31 T_ET T - Evapotranspiration (mm)
12 T_FNO3 N - Nitrate Fertilizer (mt) 32 T_QDRN T - Drain Tile Flow (mm)
13 T_FNH3 N - Ammonia Fertilizer (mt) 33 T_MUSL T - Water erosion (ton/ha)
14 T_OCPD Organic Carbon in Plow Layer (1000mt) 34 T_DRNN T - N in drain tile flow (kg/ha)
15 T_TOC Organic Carbon in Soil Profile (1000mt) 35 T_DRNP T - P in Drain Tile Flow (kg/ha)
16 T_TNO3 Total NO3 in Soil Profile (mt) 36 T_PRKP T-P in Percolation (kg/ha)
17 T_DN2 N-N2O from NO3 Denitrification (mt) 37 T_FPO T - Organic P Fertilizer (kg/ha)
18 T_YLN N Used by Crop (mt) 38 T_FPL T - Labile P Fertilizer (kg/ha)
19 T_YLP P Used by Crop (mt) 39 T_MNP T - P Mineralized (kg/ha)
20 T_FTN N Applied (mt)
*Water on agricultural lands.

The following sample script file, which is contained in the Raster Tools script directory, has all required environment variables and can be modified and run at the command line:

epicYearlyAverage4QA.csh

Other Tools and Utilities

1. Domain Grid Shapefile Generation Tool

Users can apply the domain grid shapefile generation tool to generate a polygon shapefile for a defined grid domain with the GRIDID attribute. The GRIDID attribute has values ranging from 1 for the grid cell in the lower left corner of the domain to the maximum number of cells for the grid cell in the upper right.

The following sample script file can be modified for domain shapefile generation.

generateGridShapefile.csh

2. Other Utilities

The following utility programs are stored in the SA_HOME/util directory:

  • goes_untar.pl – used to untar downloaded GOES data into the format required for the GOES cloud product processing tool.

  • updateWRFinput_landuse.R – used to update the wrfinput file using generated land use data by the NLCD and MODIS land cover generation tool. The updated wrfinput file can be used in WRF simulations with the WRF Pleim-Xiu Land Surface Model, using the 40 classes of NLCD/MODIS land cover data shown in Table 1.

Acknowledgments

The SA Raster Tools were developed with support from multiple projects:

  • Work assignments from the U.S. EPA under Contract No. EP-W-09-023, “Operation of the Center for Community Air Quality Modeling and Analysis (CMAS)”
  • NASA Research Opportunities in Space and Earth Sciences (ROSES) projects awarded to (1) the Institute for the Environment at the University of North Carolina at Chapel Hill (contract number NNX08AL28G) and (2) the National Space Science and Technology Center at the University of Alabama in Huntsville (contract number NNX09AT60G).

<< Previous Chapter - Home - Next Chapter >>

Spatial Allocator User Manual (c) 2016