Evapotranspiration Mapping in Google Earth Engine (SEBAL)

Explore evapotranspiration, combining evaporation and transpiration, and how energy and water balance influence potential and actual ET, with SEBAL mapping from satellite data.

Explore how sebal computes instantaneous evapotranspiration flux from the surface energy balance using satellite and weather data, resolving net radiation flux into latent heat and other components.

Apply the SEBAL model to map evapotranspiration from Landsat imagery, using cloud-free Landsat 8 data, header metadata, a land use map, and wind, humidity, solar radiation, and precipitation data.

Compute net radiation flux in the SEBAL framework with Landsat 8, deriving albedo, spectral radiance, reflectivity, vegetation indices, emissivities, and surface temperature to estimate shortwave and longwave fluxes.

Compute soil heat flux as the ratio to net radiation with an empirical equation, adjusting for albedo, surface temperature, and land classification.

Explain SEBAL's sensible heat flux, detailing heat transfer from surface to air via convection, using temperature differences, aerodynamic resistance, wind speed, and stability corrections with anchor pixels.

Identify cold anchor pixels from water bodies to estimate evapotranspiration and near-zero sensible heat, and hot anchor pixels in dry bare fields to map residual heat via surface temperature differences.

Explore air density and its role in calculating sensible heat flux, using the ideal gas law and standard atmospheric conditions to relate altitude, pressure, and temperature.

Highlight Landsat 8 collection 2 tile 1 row, SRTM elevation, land cover, Era5 and zero five Land hourly climate reanalysis in Google Earth Engine for evapotranspiration modeling.

Explore essential Google Earth Engine functions for evapotranspiration mapping, including image collection handling, geometry intersection, and surface albedo calculation.

Learn essential Google Earth Engine functions for SEBAL evapotranspiration mapping, including conditional expressions, surface temperature with k one and k two, dataset mapping, and sampling to feature collections.

Open your Google Earth Engine profile, create a repository for the sebal model, upload the farmland shapefile, add country shapefiles, develop the net radiation flux script, and save.

Add the Landsat 8 image collection in Google Earth Engine, filter by date, area, and cloud cover, and prepare data to compute vegetation indexes, surface albedo, and surface temperature.

Create a dataset subdivided by path and row in Google Earth Engine by filtering an image collection, mapping a function, and converting results to a list for the study area.

Create a dataset function to build image collection that computes spectral gradients, radiance, reflectivity, albedo, incoming shortwave and outgoing longwave radiation, surface emissivity, ndvi, leaf area index for evapotranspiration modeling.

Create a new dataset in Google Earth Engine, define parameters and geometry for United Arab Emirates, then compute spectral radiance for each Landsat band to support the SEBAL evapotranspiration model.

Showcases building a multi-band radiance function for eleven bands, selecting each band, multiplying by parameters, renaming to L1–L11, and returning the computed radiance image.

Compute the inverse squared relative earth-sun distance for each image, then map this distance across all bands to adjust reflectivity in the SEBAL workflow.

Compute reflectivity for each band in Landsat 8 by applying a radiance-based formula with solar atmospheric irradiance, sun elevation, and p factor, then assemble reflectivity bands.

Compute the top-of-the-atmosphere albedo using model F03 in Google Earth Engine, by selecting reflectivity bands, summing solar and atmospheric irradiance, and computing band weights to form an albedo band.

Compute surface albedo from the model using top-of-atmosphere albedo, path radiance, and atmospheric transmissivity, leveraging SRTM elevation at 30 m and clipping the study area.

Compute incoming shortwave radiation in Google Earth Engine using the solar constant, sun elevation, and atmospheric transmissivity, mapping a radiation band across the dataset with SEBAL.

Compute outgoing longwave radiation by deriving ndvi, savi, and lai from reflectance bands in Google Earth Engine, then compile these indices into a unified dataset.

Define surface emissivity as the ratio of surface to black body radiation and compute emissivities, A and B, for surface temperature and longwave radiation using ndvi and leaf area index.

Compute surface temperature with the modified Planck equation, using radiance, atmospheric terms, and modtran or radiosonde profiles. Apply Landsat eight K1 and K2 constants, convert kelvin to Celsius, remove zeros.

Compute the outgoing longwave radiation using the Stefan-Boltzmann equation in Google Earth Engine, add it as a band with this name, and save after each step in the SEBAL workflow.

Finish the new dataset function by mapping spectral radiance and reflectivity across bands, compute albedo and atmospheric transmissivity, then derive surface temperature and outgoing longwave radiation.

Debug a sebal evapotranspiration workflow in google earth engine by correcting syntax and undefined variables, then validate the United Arab Emirates agricultural areas dataset.

Identify farm locations by calculating image bounds for a Landsat image path and row, extracting a single image from a new dataset’s image collection, and using its geometry.

Identify cold and hot pixels for evapotranspiration mapping by selecting well-watered irrigation fields as cold pixels and dry harvested fields as hot pixels in Google Earth Engine (SEBAL).

Import a cold pixel location to compute incoming longwave radiation for SEBAL evapotranspiration mapping in Google Earth Engine, setting a point geometry and list with one object for later use.

Compute incoming longwave radiation in Google Earth Engine using the Egypt atmospheric emission formula, incorporating atmospheric transmissivity and minimum temperature through a mapped, multi-step dataset.

The lecture demonstrates creating a new dataset in Google Earth Engine by extracting image elements and their temperatures to prepare net radiation flux bands for evapotranspiration mapping.

Compute net radiation flux (Rn) by mapping a function over a dataset to produce a new dataset with a net radiation band, using surface albedo, shortwave, longwave, and emissivity.

Create and modify a polygon geometry in the image window to export a targeted area for evapotranspiration mapping in Google Earth Engine SEBAL.

Create a test net radiation flux image clipped to the study area in Google Earth Engine for SEBAL evapotranspiration mapping, using a 300–800 min–max and blue-to-red palette.

Map evapotranspiration with Google Earth Engine using SEBAL by computing net radiation flux for all images and dates, creating a 13-band dataset for a predefined geometry with year-month-day dates.

Export the net radiation image to Google Earth Engine assets and Google Drive, converting to float32, setting a 30 m scale, and using date-based naming for later evapotranspiration modeling.

Export net radiation flux to Google Drive, download the TIFF rasters, and open them in ArcGIS or QGIS to analyze albedo, NDVI, SAVI, and temperature for evapotranspiration mapping.

Develop momentum roughness length for sensible heat flux within a Google Earth Engine workflow, using Copernicus land use land cover classification at 100 m to extract UAE class areas.

Compute momentum roughness length for the United Arab Emirates using Copernicus land cover classes, with agriculture adjusted by leaf area index and visualized via a palette.

Sample land cover from a shape file at 100 m resolution to assign each pixel a land cover class, then enumerate four classes (20, 40, 50, 60) for evapotranspiration mapping.

Verify the pixel count for momentum reference length by computing class frequencies, then export the classified image to an asset or drive, resampling 100 m to 30 m for Landsat.

Import uncorrected momentum roughness lengths and net radiation flux into the Sebal script, define and clip study-area geometry with Copernicus land cover, and layer leaf area index for visualization.

Correct the momentum roughness length for agricultural fields by applying LAI values, visualize the corrected results, and export the image as an asset with a scale factor of 30.

Import hot and cold pixels, roughness length and farm shapefile, define coordinates as point geometries, and prepare Landsat 8 data to compute sensible and solar heat flux in SEBAL.

Explore visualization parameters for evapotranspiration mapping in Google Earth Engine using sebal, covering net radiation, momentum roughness length, ndvi, surface temperature, albedo, and soil heat flux with their color scales.

Define bands to sample from image variable, selecting net radiation flux, surface temperature, and surface albedo using the first band, and create a geometry variable for the study area.

Create and compare false-color vegetation and agriculture layers in Google Earth Engine by selecting multispectral bands, adjusting visualization parameters, and highlighting cold versus hot pixels.

Compute the soil heat flux to net radiation ratio using a formula involving surface radiation, albedo, and NDVI, and name the resulting image grn_ratio.

Compute the grain ratio from ndvi, setting it to 0.5 if ndvi is negative, then derive soil heat flux G as grain ratio times net radiation, and compute Rn ratio.

Add layers using the visualization parameters for net surface radiation, surface albedo, and soil heat flux to net radiation ratio; adjust surface temperature and soil heat flux ranges, then run.

Sample cold and hot pixels for SEBAL by extracting end band, momentum roughness length, and surface temperature, converting to kelvin, and comparing net radiation with soil heat flux and evapotranspiration.

Shows a practical example computing momentum roughness length, friction velocity at 2 m, wind speed at 200 m, and aerodynamic resistance to heat transport for sensible heat flux in SEBAL.

Calculate momentum roughness length for a weather station using Carmen's constant and vegetation height near 0.3 m, with site vegetation at least 3.3 m, and report 2 m wind speed.

Compute wind velocity at 200 m above weather station using friction velocity at 2 m, log 200 over roughness length, and Carmen's constant, then compare to wind at 2 m.

Learn to compute air density for hot and cold pixels using surface pressure and temperature at two metres to estimate sensible and latent heat flux for evapotranspiration in SEBAL.

Export evapotranspiration results from a Google Earth Engine script to csv by creating a feature collection with overridden metadata for cold and hot pixels, then exporting to Google Drive.

Explain the iterative process for calculating sensible heat flux in SEBAL, including wind speed, friction velocity, aerodynamic resistance, temperature differences, and stability corrections, repeated until stabilization.

Discuss the hourly weather inputs for reference evapotranspiration in Google Earth Engine: 2 m temperatures, hourly surface net solar radiation, dew point, wind, and day and hour.

Add the ECMWF climate reanalysis collection to derive weather parameters for reference evapotranspiration in a Google Earth Engine SEBAL workflow, using weather station data, climate variables, and 24-hour image extraction.

Select dew point and temperature at two meters, ten-meter wind components, and surface net solar radiation, clip to the study area, and save the 24-hour hourly dataset as a list.

Compute day of year, hour and minutes, temperature in kelvin, surface net solar radiation hourly, wind components, and dew point for evapotranspiration mapping in Google Earth Engine SEBAL.

Map a function over an image collection to create a feature collection, computing weather parameters and including day of year, year, minutes, temperature, surface radiation, wind, and dew point.

Convert the image collection to a feature collection, obtain the weather station elevation from SRTM data in the UAE, and export climate data to a CSV in Google Drive.

Compute reference evapotranspiration for a given day by mapping weather parameters to eight inputs in the University of Idaho software, export results, and integrate into an Excel workflow.

Explain how to convert GMT to local time for satellite overpasses, select appropriate hourly weather data to compute instantaneous wind speed and reference evapotranspiration for SEBAL.

Develop a script to compute instantaneous wind speed and instantaneous transpiration for a Landsat image, aligning with the image date and GMT scene center time, in Google Earth Engine.

Develop a script that converts string times to numbers, applies GMT/UTC corrections, and computes local hour and time delta for instantaneous wind and ETr calculations.

Learn to compute instantaneous wind speed and instantaneous reference evapotranspiration with a QGIS plugin, and derive daily evapotranspiration for the image day using weather data.

Use the GetSebal plugin via the GitHub plugin workflow to compute correlation coefficients for temperature differences, fill export data from CSV, and save A and B for the evapotranspiration model.

Compute daily evapotranspiration with SEBAL in Google Earth Engine by importing net radiation flux, wind speed, soil heat flux, and Landsat imagery, then derive correlations, boundaries, and ndvi-based estimates.

Finish Google Earth Engine by importing von Karman constants and vegetation height, computing momentum roughness length for weather station, cold blue and hot yellow pixels, air density.

Compute temperature difference (DT) across the image to derive sensible heat flux and instantaneous evapotranspiration (latent heat flux) for SEBAL in Google Earth Engine, including friction velocity and aerodynamic resistance.

Compute instantaneous evapotranspiration and the transpiration fraction, then derive daily evapotranspiration using SEBAL in Google Earth Engine. Visualize results with color palettes, export maps, and repeat the workflow across images.

Re-computes the momentum roughness length corrected for January 13 to enable applying the SEBAL model to all yearly images using the existing scripts.

Recompute soil heat flux for an image date in Google Earth Engine, updating net radiation flux, momentum roughness length, and weather data to derive day-of-year hot and cold pixel values.

Export hourly weather parameters for a new image in google earth engine, updating the date and export name to save climate data for January 13.

Compute hourly reference evapotranspiration (ETr) in the SEBAL workflow by preparing eight climate fields (day of year, dew point, hours, minutes, Tmax, Tmin, wind, year) and exporting a climate CSV.

Compute instantaneous wind speed and instantaneous reference evapotranspiration (ETr) for the new image using the updated t1-t2 script and the 11:00–12:00 time window.

Use the jezebel plugin in qgis to input cold and hot pixel values, wind speed at 200 m, and transpiration, then save A and B correlation coefficients for temperature difference.

Calculate SEBAL-based evapotranspiration for January 13 in Google Earth Engine using correlation coefficients for temperature difference and instantaneous wind speed, plus instantaneous reference for transpiration.