OpenET - geeSEBAL

Estimating Evapotranspiration using SEBAL model in Google Earth Engine platform.

• The Google Earth Engine Surface Energy Balance for Land (geeSEBAL) solves the energy balance equation (LE + H = Rn - G) to estimate Daily Evapotranspiration (ET) by using Landsat images (L4, L5, L7, L8, and L9) and meteorological data (air temperature, relative humidity, global radiation and wind speed).

Input Collections

• The following Earth Engine image collection are use in geeSEBAL:

Image Collections IDs
LANDSAT/LC09/C02/T1_L2
LANDSAT/LC08/C02/T1_L2
LANDSAT/LE07/C02/T1_L2
LANDSAT/LT05/C02/T1_L2
LANDSAT/LT04/C02/T1_L2

Model Description

• Surface Energy Balance Algorithm for Land (SEBAL) was developed and validated by Bastiaanssen (Bastiaanssen et al., 1998a, 1998b) to estimate evapotranspiration (ET) from energy balance equation (Rn – G = LE + H), where LE, Rn, G and H are Latent Heat Flux, Net Radiation, Soil Heat Flux and Sensible Heat Flux, respectively. SEBAL estimates LE as a residual of others energy fluxes (LE = Rn - LE - G).
• SEBAL algorithm has an internal calibration, assuming a linear relationship between dT and LST across domain area, where dT is designed as a vertical air temperature (Ta) floating over the land surface, considering two extreme conditions. At the hot and dry extreme condition, LE is zero and H is equal to the available energy, whereas at the cold and wet extreme condition, H is zero and LE is equal to the available energy.
• Workflow of geeSEBAL, demonstrating remote sensing and global meteorological inputs, as well as data processing to estimate daily evapotranspiration.

Model Design

Image()

• Compute Daily ET or ET fraction for a single input image.
• Allow to obtain ET image collections by mapping over Landsat collections.

Landsat Collection 2 Input Image

• Select Image.from_landsat_c2_sr() method to instantiate the class for a Landsat Collection 2 SR image. Image must have the following bands and properties:

SPACECRAFT_ID	Band Names
LANDSAT_4	SR_B1, SR_B2, SR_B3, SR_B4, SR_B5, SR_B7, ST_B6, QA_PIXEL
LANDSAT_5	SR_B1, SR_B2, SR_B3, SR_B4, SR_B5, SR_B7, ST_B6, QA_PIXEL
LANDSAT_7	SR_B1, SR_B2, SR_B3, SR_B4, SR_B5, SR_B7, ST_B6, QA_PIXEL
LANDSAT_8	SR_B1, SR_B2, SR_B3, SR_B4, SR_B5, SR_B6, SR_B7, ST_B10, QA_PIXEL
LANDSAT_9	SR_B1, SR_B2, SR_B3, SR_B4, SR_B5, SR_B6, SR_B7, ST_B10, QA_PIXEL

PROPERTIES
system: index - Landsat scene ID (ex: LC08_044033_20170801)
system: time_start - Time start of the image in epoch time
SPACECRAFT_ID - Landsat Satellite (LANDSAT_4, LANDSAT_5, LANDSAT_7, LANDSAT_8, LANDSAT_9)
SUN_ELEVATION - Solar elevation angle in degrees

Model Output

The general outputs of the geeSEBAL are ndvi (normalized difference vegetation index), lst (land surface temperature), et_fraction and et. They can be selected as example below:

Example

import openet.geesebal as geesebal

ls_img = ee.Image('LANDSAT/LC08/C02/T1_L2/LC08_044033_20170801')
model_obj = geesebal.from_landsat_c2_sr(ls_img)

ndvi = model_obj.ndvi
lst = model_obj.lst
et_fraction = model.et_fraction
et = model_obj.et

Examples Notebooks

Examples of how to use geeSEBAL model are detailed in examples folder:

geeSEBAL examples.

Installation

pip install openet-geesebal

Depedencies

• earthengine-api https://github.com/google/earthengine-api`
• openet-core https://github.com/Open-ET/openet-core`

References

[2021] Laipelt, L., Kayser, R. H. B., Fleischmann A., Ruhoff, A., Bastiaanssen, W., Erickson, T., Melton, F. Long-term monitoring of evapotranspiration using the SEBAL algorithm and Google Earth Engine cloud computing.