Lesson 3 of 8 · New Zealand–NASA Earth observation
New Zealand–NASA research: turn observations into useful environmental evidence
8 minute lesson · 3 checks along the way
Your goal for this lessonExplain why independent validation and local knowledge matter to Earth-observation projects.
In this lesson 3 sections · 3 checks
Using archive links & definitions
Follow a linked name to explore its archive card. “Back to your lesson” brings you to the same part of the reading.
Words with a dotted underline open a short definition right here in the lesson.
Read & understand
From feasibility studies to shared research
The New Zealand–NASA Earth-observation research program followed a 2022 cooperation framework. Twelve feasibility studies began in 2023, and five larger collaborative projects were supported from June 2025. Their subjects include forests, coastal waters and water systems. These projects contribute research methods and regional observations; they do not imply that New Zealand owns every NASA satellite supplying the data.
A project begins with a question that a measurement can help answer. Forest managers might need information about growth or disturbance, while coastal researchers may need to distinguish suspended sediment from biological activity. The useful product is an interpretation with a known scope, not merely an attractive image. Choosing the sensor and testing the method should follow from the environmental question.
Check 1 of 3
10 pointsUse the reading above. You can try again without losing points.
Read & understand
Color is evidence, not a complete diagnosis
The Landsat 8 observation in this lesson shows the Canterbury coast on 19 May 2021, before the later joint projects. NASA’s processing emphasizes water color. The accompanying interpretation describes sediment transport and chlorophyll-related effects. Those different contributors help explain why a green or turquoise pixel cannot by itself supply a single, certain water-quality diagnosis.
Imagine a model estimating a water property from image color. The team should compare its estimates with suitable measurements taken near the same locations and times. A sample collected months later may describe a different condition. Clouds, viewing geometry, atmospheric effects and mixed shoreline pixels can also influence the comparison. State what the model estimates and how the reference measurements support that interpretation.
Landsat 8 observation acquired on 19 May 2021. NASA applied color filtering to emphasize coastal detail. This historical environmental context predates the later New Zealand–NASA research projects and is not presented as one of their results. Original publisher image preserved; display resized and converted to WebP.
Look closely
Can a turquoise or green area alone identify one particular water-quality cause?
See what to notice
No. NASA’s interpretation identifies sediment and chlorophyll-related contributions. The processed image supplies spatial evidence, while quantitative interpretation needs suitable measurements and a validated method.
Check 2 of 3
10 pointsUse the reading above. You can try again without losing points.
Read & understand
Evaluate where the method will be used
In an invented set, predicted values are twelve, eighteen and twenty-six units; independent measurements are ten, twenty and twenty-five. The absolute errors are two, two and one, giving a of five divided by three, about 1.67 units. The example measures average disagreement for those three cases. It does not establish the largest future error or performance everywhere in New Zealand.
A useful evaluation should include the conditions important to the intended users: different seasons, landscapes and measurement ranges. Local knowledge can help identify missing conditions and meaningful questions, while communities should help shape how relevant information is interpreted and used. Shared research becomes valuable when its methods, uncertainties and data context travel with the result, allowing another team to assess whether it fits their own decision.
Check predictions against independent measurements
Original teaching diagram. INVENTED ERRORS. Absolute errors: 2, 2, 1. Total error: 5 units. MEAN ABSOLUTE ERROR. 5 ÷ 3 ≈ 1.67 units. Applies to this test sample. A sample average is not a guarantee for every season, location or future observation.
Open full-size diagram in a new tab: Check predictions against independent measurementsCheck 3 of 3
10 pointsUse the reading above. You can try again without losing points.
Put it together
Your lesson progress
Each check connects to a part of the story. Complete all 3 to earn your 25-point lesson bonus.
Explore further · Optional
Go to the source
Explore the evidence behind this lesson. These optional readings and videos open in a new tab.
- Read: MBIE: New Zealand–NASA Earth-observation researchDistinguishes 2023 feasibility studies from five larger projects beginning in June 2025. Project objectives are not all treated as completed results.
- Read: MBIE: international space research projectsMaps the NASA, DLR and Australian collaborations. Use the detailed program records for dates and individual demonstrations.
- Read: NASA Earth Observatory: Coastal Color in New ZealandThe 19 May 2021 Landsat 8 image is processed to emphasize coastal color. Sediment and chlorophyll are different contributors; this is context predating the later research program.