How to get Planet satellite data

I you never heard about Planet, please read the Medium post below where I introduce the PlanetScope satellite data from a technical perspective.

Planet(Scope) satellite imagery: a technical introduction

In this post, I will guide you through the process of getting your hands on the actual Planet data.

Example of a Planet image. Source: Planet Explorer user guide

Step 1: Creating a Planet account

There are several options for creating a Planet account, the preferred one depends on your data needs and your affiliation.

• Education and Research Program: any university-affiliated student, faculty member or researcher may apply for this. It provides limited, non-commercial access to PlanetScope and RapidEye imagery. However, a university email address is required. https://www.planet.com/markets/education-and-research/
• ESA Earthnet Program: Planet is officially registered as a Third Party Mission in the European Space Agency’s EarthNet program. This allows European scientists, researchers, and companies to apply for data through the ESA platform. https://www.planet.com/earthnet/
• NICFI Satellite Data Program: Funded by the Norwegian International Climate & Forests Initiative (NICFI), this program provides PlanetScope Visual and Surface Reflectance (Analysis Ready) Basemaps for all forested areas in the tropics. A disctinction is made between historical data from December 2015 to August 2020 (one basemap every six months) and recent data from September 2020 to August 2022 (one basemap for each month). https://www.planet.com/nicfi/ More info about (the use of) these data can be found here: https://university.planet.com/page/nicfi-resources
• Nonprofit Program: Planet offers small and medium data bundles to non-profit organizations worldwide. The price of these are not disclosed, which is unfortunate. https://www.planet.com/nonprofit/

Once you have created an account, you can log into the dashboard to see your monthly quota and your actual usage on https://www.planet.com/account/#/plans

If, like me, you are in the Education and Research Program, you have a personal quota of 5,000 km² per month. Luckily, as I will explain in the next step, you can download data clipped out to your exact region(s) of interest, limiting your usage to what you really need.

Step 3: Finding and filtering Planet data

From the dashboard, there is a direct link to the Planet Explorer, the portal where you can find and download Planet data: https://www.planet.com/explorer/

Setting your Area Of Interest (AOI)

In Planet Explorer, you can search for satellite acquisitions over your region of interest. In the search field (top left), you can search aby municipality/city or by street name. Some other place types (like forests) are also findable, but not in a systematic way.

Once you have zoomed to the rightregion, you can delineate your exact Area Of Interest (AOI). You can find this option in the lower right corner of the Explorer (see figure below).

Delineating an Area Of Interest (AOI)

You can choose between different shapes for the AOI, including a circle, a rectangle or a free shape. Importing geodata files as AOI is also an option, here you can choose between some of the most common geospatial vector formats (e.g. shp, (geo)json, kml). If your geodata file contains multiple polygons, you can select each one separately (or combined). You can easily download your AOI as a geojson file for future reference.

Your images will be clipped to the exact shape of your AOI before download and the size (in km²) indicated after delineation is the size that will be subtracted from your quota if you order an image for this AOI. Note: if you order multiple acquisitions over the same AOI, the size will be multiplied by the number of acquisitions.

Filtering acquisitions

Once your AOI is set, you will see a list of acquisitions appear on the left side of the screen. Below the map, you can also see a timeline that indicates the dates with acquisitions.

Overview of available acquisitions

However, as you can see in the figure above, not all acquisitions have good data. To find the acquisitions that fulfill your requirements, you can use the filter (top right of the screen).

There you can select:

• imagery type: High Resolution, Medium Resolution or Low Resolution. The default choice is Medium Resolution, as this includes the native Planet(Scope) data.
• spectral bands: “NIR” (= only 4-band or 8-band images) or “coastal blue, green I, yellow, red edge” (= only 8-band images)
• instrument type: the different types of Doves. PS2 and PS2.SD have 4 bands, while PSB.SD has 8 bands.
• publishing stage: always select Standard & Finalized.
• area coverage: sliders to select minimum and maximum coverage of the area. Here, especially the minimum is important as it indicates the percentage of your AOI covered by an image.
• cloud cover: sliders to select minimum and maximum cloud coverage of the image. Here, especially the maximum is important as it indicates the percentage of the entire image (thus not the AOI) covered by clouds.

You can also check the ‘include only surface reflectance’ and ‘show only standard image quality’ options at the bottom to make sure your images are atmospherically corrected and have good quality.

In this example, we filtered for 8-band PlanetScope scenes with a minimum area coverage to 90% and a maximum cloud cover to 5%. We can immediately see from the list of acquisitions and from the timeline that the number of images after filtering is markedly lower than the number before filtering.

Overview of available acquisitions after filtering

Step 4: Ordering and downloading Planet data

If you click on an acquisition in the list, the eye icon becomes blue and you can see its coverage on the map. The image remains blurred, since you have not yet secured access.

You can see that for some dates multiple scenes are available. If you click on the blue box with the number of scenes, you get more details about each of these images.

You can also visually check the thumbnail for each image/scene, to assess its quality. Below, you can see a hazy image thumbnail versus a clear one.

Visually assessing the quality of the thumbnail pictures

Once you have decided on the scenes that you want to download, you can click the shopping cart button to add them to your order. Note: if you click the shopping cart before going to the separate scenes, all scenes from that date are automatically added to your order.

You can finalize your order by clicking the ‘order scenes’ button at the bottom of the list of acquisitions. The number of scenes in the order is indicated between brackets. Please double check that you have only added the scenes you really want to order!

Ordering Planet data

By default, ‘direct download’ is selected as the delivery destination. If you want to add the ordered images to your hosted data for analysis, you can check that option instead. Read more about this here.

There are three options for the file format: GeoTiff, Cloud Optimized GeoTiff and You can select Cloud Optimized GeoTif if you want to host the data in the cloud. NITF 2.1 is a format developed by the US government for easy data transfer, but is not a standard for most applications.

The next step in the ordering process is setting the spectral data requirements. Here you can choose between 3-band (RGB), 4-band (RGB-NIR) and 8-band data; and between Top Of Atmosphere radiance (not atmospherically corrected) or Bottom Of Atmosphere surface reflectance (atmospherically corrected). For terrrestrial applications, we usually select surface reflectance.

Setting the spetral data requirements

In the final step, you have to specify whether you want to order the entire scene or only the part clipped by your AOI. To save data quota, it is best to clip the items to AOI. You can choose to composite all items to a single Geotiff or keep each item as a separate file. There is also an option to harmonize the spectral characteristics of the PlanetScope images to Sentinel-2. According to this white paper, the scene level normalization tool significantly reduces variability between scenes and between generations of PlanetScope satellites while maintaining reasonable radiometric accuracy and compatibility with Sentinel2 data. As I have not yet experimented with the harmonization, I cannot make any recommendations about this myself.

The final tweaking steps

After you click ‘order’, it can take some time before the data become available. You will receive a notification by email when this happens. You can also check the availability of all your orders by activating the ‘view and download past order’ tab on the left of the screen.

I hope you enjoy the use of the PlanetScope data for your research or educational purposes!