> For the complete documentation index, see [llms.txt](https://docs.geopard.tech/geopard-tutorials/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.geopard.tech/geopard-tutorials/product-tour-web-app/zones-maps-and-analytics/multi-layer-analytics.md).

# Multi-Layer Analytics

With GeoPard, you can integrate multiple data layers to create management zones. You control weights, indices, the number of zones, clustering type, and minimum polygon area.

{% @arcade/embed url="<https://app.arcade.software/share/em63luvNTYI6CFKUC3m6>" flowId="em63luvNTYI6CFKUC3m6" %}

*Interactive demo - multi-layer zones creating*

Precision agriculture generates large amounts of data, including yield data, satellite imagery, and soil fertility layers. Many teams still struggle to turn those layers into useful insights and actions. In precision agriculture, management zones are parts of a field with similar yield potential based on soil type, slope position, soil chemistry, microclimate, or other crop-driving factors. Local field knowledge is also an important part of the process. Management zones help optimize crop inputs and yield potential.

The main challenge is building management zones that reflect real field variability. Combining layers such as satellite imagery, soil fertility, topography derivatives, and yield monitor data is the next logical step toward more responsive management zones.

Multi-layer analytics (also known as integrated analysis) is becoming a part of the GeoPard geospatial analytics engine.

Classic combinations for integrated analysis include one or more yield datasets, an NDVI map, elevation, and soil sensor physicochemical characteristics. GeoPard supports these parameters and also lets you include other field data layers already available in the system or uploaded directly by the user, such as soil sampling and yield datasets. As a result, you can work with a full set of parameters for integrated analytics:

Yield data\
Remote sensing data:

* Potential productivity map (single-year and multi-year)
* Stability/variation map
* Vegetation indices NDVI, EVI2, WDRVI, LAI, SAVI, OSAVI, GCI, GNDVI

Topography:

* Elevation
* Slope
* Curvature
* Wetness index
* Aspect
* Hillshades
* Ruggedness
* Roughness

Soil data:

* pH
* CEC (cation exchange capacity)
* SOM (soil organic matter)
* K (potassium)
* Thin topsoil depth, lower available water holding capacity (drought-prone soil)
* EC (electrical conductivity)
* and other chemical attributes available in the uploaded dataset

It’s important to note that you can configure custom factors for every data layer to assign the desired weight. You can build management zone maps based on your knowledge of the field while selecting data sources and their weights in GeoPard.

## How layer weights work

Weight sets how much each layer influences every pixel of the resulting zones map.

### Normalization comes first

Input layers use incompatible units and ranges: NDVI runs 0 to 1, yield 4 to 12 t/ha, pH 5.5 to 7.5, EC 5 to 60 mS/m. Combining raw values would let whichever layer has the largest numeric range dominate the result regardless of its agronomic importance.

GeoPard applies min-max normalization to each layer first, rescaling it to a common range based on that layer's own minimum and maximum within the field. Weight alone then controls influence.

### Then weights apply

Each normalized layer is multiplied by its weight and the layers are combined into a single surface.

### Negative weights invert the layer

A negative weight flips the scale: the layer's high values are treated as low and its low values as high. Use this when high readings indicate worse conditions, for example:

* Clay content on a poorly drained field
* Elevation where low spots suffer frost or waterlogging
* EC where high readings indicate salinity rather than water-holding capacity

### Gaps are interpolated

Where a layer has missing pixels (cloud cover on a satellite scene, a skipped pass in yield data, sparse soil sample points), GeoPard interpolates the gap from surrounding values rather than dropping the pixel. The combined map stays continuous across the whole field.

### The result is normalized again

After weighting and combining, the result map is normalized once more so it spans the full range, then split into the number of zones you selected.

{% hint style="info" %}
**Choosing weights:** start by giving your most trusted layer the highest weight. Multi-year yield and long-term satellite productivity are usually the most reliable signals. Single-date satellite imagery is the most situational, since one scene captures one moment in the season.
{% endhint %}

The images below show a sample field with data layers such as an 18-year productivity map, digital elevation model, slope, hillshade, and 2019 yield data, along with different integrated analytics outputs. You can follow how the management zones evolve as additional layers are added.

<figure><img src="https://3272281156-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FYICBELdyAXXebKAzfLOR%2Fuploads%2Fthi0JMlu2pSoF0wLOJp4%2FGeoPard%20Multi-Layer%20Maps%20creating.png?alt=media&amp;token=05cd9cea-0069-451d-af90-caa61801aaa1" alt="Multi-layer zones setup with several data layers and weights"><figcaption><p>Layers and weights for a multi-layer zones map.</p></figcaption></figure>

<figure><img src="https://3272281156-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FYICBELdyAXXebKAzfLOR%2Fuploads%2FrfxM1lvXMqXgHLGug8Ow%2FGeoPard%20Multi-Layer%20Maps%20creating%20-%20result%20map.png?alt=media&amp;token=78d3bfd6-255b-477b-95d5-6bb0de84da61" alt="Multi-layer zones map built from imagery, soil EC, topography and yield"><figcaption><p>The result: zones from imagery, soil EC, topography and yield together.</p></figcaption></figure>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.geopard.tech/geopard-tutorials/product-tour-web-app/zones-maps-and-analytics/multi-layer-analytics.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
