FAQ

Frequently Asked Questions

How do reductions behave with partial pixels?

The weights of pixels used during an aggregation process are based on the overlap between the smaller pixels being aggregated and the larger pixels specified by the output projection. This is illustrated in the figure below.

The default behavior is that input pixel weights are computed as the fraction of the output pixel area covered by the input pixel. In the diagram, the output pixel has area a, the weight of the input pixel with intersection area b is computed as b/a and the weight of the input pixel with intersection area c is computed as c/a. 

This behavior can result in unexpected results when using a reducer other than the mean reducer. For example, to compute forested area per pixel, use the mean reducer to compute the fraction of a pixel covered, then multiply by area (instead of computing areas in the smaller pixels then adding them up with the sum reducer).

What do the different variable names stand for?

ET: Evapotranspiration 
ET_MAD_MIN: Evapotranspiration Median Absolute Deviation Minimum
ET_MAD_MAX: Evapotranspiration Median Absolute Deviation Maximum
ETo: Grass Reference Evapotranspiration
ETr: Alfalfa Reference Evapotranspiration
ETof: Fraction of Reference Evapotranspiration
NDVI: Normalized Difference Vegetation Index 
PR: Precipitation 
COUNT: Number of Cloud Free Landsat observations 
MODEL_COUNT: Number of model included in observation for Ensemble query

What are the requirements for an asset ID?

Each asset must be shared with READ permission to OpenET (See Earth Engine Tab). Each geometry in your shapefile must be simple and non-overlapping. Each polygon must contain a unique identifier and your shapefile must conform to the limits of your account (See Quota).

Can I export to my own Google Bucket? (BUG FIX PENDING)

CURRENTLY NOT WORKING

Yes, but you will need to do a few things first. When you create you bucket, it MUST be public.

You will also need to click the 3 dots to the right of you bucket and use the +ADD PRINCIPAL button. In the new principals box type: openet-data-explorer-raster-qu@openet.iam.gserviceaccount.com. Under the ASSIGN ROLES section choose STORAGE LEGACY BUCKET WRITER and hit save. Now for export based queries, if you pass the bucket parameter with the name of your bucket exports will be directed there.

For daily overpass collections, what models have what native variables?

Ensemble: et
SIMS: etof
DisAlexi: et
eeMETRIC: etof
geeSEBAL: et
PTJPL: et
SSEBop: etof

What happens if I exceed the request limits?

Queries which attempt to exceed the limits of the account's quota will be blocked. If you have exceeded your requests per month, you must wait until the following 1st of the month (UTC time) query the API again.

What data is considered provisional?

Realtime OpenET data is considered “provisional” for the last 120 days. This data can and will change and is not considered stable/static/final. The two main sources for this change are the gridMET data not being finalized until after 60 days, and new Landsat images needing to be run through the models and used to interpolate previous months. To account for all of this, we make monthly updates of the last 3 to 4 month for all models and the ensemble.

I linked my Earth Engine account, does OpenET now have access to my whole Google Drive?

OpenET can only see, edit, create, and delete only the specific Google Drive files you use with our API. This access can be revoked by you at anytime by access your Google account settings.

Using the parameter overpass=true returned an error. What does this mean?

The overpass flag only works for certain model and variable combinations at this time. The satellite overpass data varies for each model and passing overpass=true will only work with the native band of the model. Ensemble, geeSEBAL, DisALEXI, and PTJPL support ET as the native band while eeMETRIC SSEBop, and SIMS support ETof directly.

What do I do if I receive the Invalid API Credentials error message?

This error can occur for a few different reasons. The most common being a miss-typed or expired api key. Remember, when querying the api you will need to use the "Authorization" header format as shown in the code examples. If you have ruled out the previous three possibilities retry the query after a few seconds as the system may be experiencing high volumes of traffic.

If you api key has expired you will need to generate a new one from you project dashboard.

Do field boundaries in the geodatabase change year to year?

Field boundaries generally remain static, however priority updates are ingested to include more recent field boundary datasets provided by each state if they include significant revisions.

What spatial reference system is used for raster geoTIFF exports?

EPSG:4326

What is a C.O.G.?

A Cloud Optimized GeoTIFF (COG) is a regular GeoTIFF file aimed at being hosted on a HTTP file server with an internal organization that enables more efficient workflows on the cloud. It does this by leveraging the ability of clients issuing HTTP GET range requests to ask for just the parts of a file they need.

What is an Earth Engine Compute Unit (EECU)

An Earth Engine Compute Unit (EECU) is a mechanism for representing an amount of instantaneous processing power. OpenET tracks the total computational footprint of tasks as a function of their EECU usage through time (EECU-seconds).

What is the largest response the API can return?

The OpenET API is limited to a 32mb maximum response size. Any response which exceeds this size may be returned as a compressed zip file. See specific endpoint documenation for more details.

When downloading a raster url return I get the error "Parameter 'left' is required."

This error may occur if the geometry used fully overlaps a portion of masked pixels within a raster. This can most frequently happens with the SIMS model as it is masked for irrigated agriculture only. Remove any polygon ID listed in the error message and retry the query. An example of what this looks is shown below.

This can also occur with overlapping polygon geoemtries in your shapefile. You will need simple, non-overlapping geometries for all OpenET endpoints.

Where can I find more information about OpenET's models?

The OpenET website's methodologies page: https://openetdata.org/methodologies will be yout best resource here.

When I use the drive_folder parameter multiple destination folders get created. What can I do to solve this?

This is a know bug with Earth Engine when exporting to Google Drive. You can mitigate this issue by creating the export folder in the root of your Google Drive in advance.

How do I cite OpenET data?

We have citations available for each model in our Earth Engine collections. Here is a link for the Ensemble v2_0 collection: 

https://developers.google.com/earth-engine/datasets/catalog/OpenET_ENSEMBLE_CONUS_GRIDMET_MONTHLY_v2_0#citations

What is the latency associated with the OpenET daily and monthly collections?

Each data product used to generate OpenET data has the potential to introduce additional latency. In addition to the meteorological and ancillary data inputs, OpenET relies on the Landsat system of satellites. Current data production uses imagery from Landsat 8 & 9. Each satellite makes a complete orbit every 99 minutes, completes about 14 full orbits each day, and crosses every point on Earth once every 16 days. The satellite orbits are offset to allow 8-day repeat coverage of any Landsat scene area on the globe. Coverage may change depending on the regions you are looking to retrieve data. OpenET does mask for clouds, which during rainy months may add additional latency between usable observations.

Provisional monthly data typically becomes available on the 5th or 6th of the following month. However, this may vary if there is a high number of cloudy scene for the region of interest. 

Using the OpenET raster/metadata endpoint, you can retrieve the date_ingested parameter which will provide the most recent time data was added for a specific region.

Don't see your question here? Contact us.

Last updated 1 month ago

The weights of pixels used during an aggregation process are based on the overlap between the smaller pixels being aggregated and the larger pixels specified by the output projection. This is illustrated in the figure below.

The default behavior is that input pixel weights are computed as the fraction of the output pixel area covered by the input pixel. In the diagram, the output pixel has area a, the weight of the input pixel with intersection area b is computed as b/a and the weight of the input pixel with intersection area c is computed as c/a. This behavior can result in unexpected results when using a reducer other than the mean reducer. For example, to compute forested area per pixel, use the mean reducer to compute the fraction of a pixel covered, then multiply by area (instead of computing areas in the smaller pixels then adding them up with the sum reducer).

ET: Evapotranspiration ET_MAD_MIN: Evapotranspiration Median Absolute Deviation Minimum ET_MAD_MAX: Evapotranspiration Median Absolute Deviation Maximum ETo: Grass Reference Evapotranspiration ETr: Alfalfa Reference Evapotranspiration ETof: Fraction of Reference Evapotranspiration NDVI: Normalized Difference Vegetation Index PR: Precipitation COUNT: Number of Cloud Free Landsat observations MODEL_COUNT: Number of model included in observation for Ensemble query

Each asset must be shared with READ permission to OpenET (See Earth Engine Tab). Each geometry in your shapefile must be simple and non-overlapping. Each polygon must contain a unique identifier and your shapefile must conform to the limits of your account (See Quota).

CURRENTLY NOT WORKING Yes, but you will need to do a few things first. When you create you bucket, it MUST be public. You will also need to click the 3 dots to the right of you bucket and use the +ADD PRINCIPAL button. In the new principals box type: openet-data-explorer-raster-qu@openet.iam.gserviceaccount.com. Under the ASSIGN ROLES section choose STORAGE LEGACY BUCKET WRITER and hit save. Now for export based queries, if you pass the bucket parameter with the name of your bucket exports will be directed there.

Queries which attempt to exceed the limits of the account's quota will be blocked. If you have exceeded your requests per month, you must wait until the following 1st of the month (UTC time) query the API again.

Realtime OpenET data is considered “provisional” for the last 120 days. This data can and will change and is not considered stable/static/final. The two main sources for this change are the gridMET data not being finalized until after 60 days, and new Landsat images needing to be run through the models and used to interpolate previous months. To account for all of this, we make monthly updates of the last 3 to 4 month for all models and the ensemble.

The overpass flag only works for certain model and variable combinations at this time. The satellite overpass data varies for each model and passing overpass=true will only work with the native band of the model. Ensemble, geeSEBAL, DisALEXI, and PTJPL support ET as the native band while eeMETRIC SSEBop, and SIMS support ETof directly.

This error can occur for a few different reasons. The most common being a miss-typed or expired api key. Remember, when querying the api you will need to use the "Authorization" header format as shown in the code examples. If you have ruled out the previous three possibilities retry the query after a few seconds as the system may be experiencing high volumes of traffic. If you api key has expired you will need to generate a new one from you project dashboard.

A Cloud Optimized GeoTIFF (COG) is a regular GeoTIFF file aimed at being hosted on a HTTP file server with an internal organization that enables more efficient workflows on the cloud. It does this by leveraging the ability of clients issuing HTTP GET range requests to ask for just the parts of a file they need.

An Earth Engine Compute Unit (EECU) is a mechanism for representing an amount of instantaneous processing power. OpenET tracks the total computational footprint of tasks as a function of their EECU usage through time (EECU-seconds).

This error may occur if the geometry used fully overlaps a portion of masked pixels within a raster. This can most frequently happens with the SIMS model as it is masked for irrigated agriculture only. Remove any polygon ID listed in the error message and retry the query. An example of what this looks is shown below.

We have citations available for each model in our Earth Engine collections. Here is a link for the Ensemble v2_0 collection: https://developers.google.com/earth-engine/datasets/catalog/OpenET_ENSEMBLE_CONUS_GRIDMET_MONTHLY_v2_0#citations

Each data product used to generate OpenET data has the potential to introduce additional latency. In addition to the meteorological and ancillary data inputs, OpenET relies on the Landsat system of satellites. Current data production uses imagery from Landsat 8 & 9. Each satellite makes a complete orbit every 99 minutes, completes about 14 full orbits each day, and crosses every point on Earth once every 16 days. The satellite orbits are offset to allow 8-day repeat coverage of any Landsat scene area on the globe. Coverage may change depending on the regions you are looking to retrieve data. OpenET does mask for clouds, which during rainy months may add additional latency between usable observations. Provisional monthly data typically becomes available on the 5th or 6th of the following month. However, this may vary if there is a high number of cloudy scene for the region of interest. Using the OpenET raster/metadata endpoint, you can retrieve the date_ingested parameter which will provide the most recent time data was added for a specific region.