Currently the Climate API is not versioned and is under active development - breaking changes should be expected, and backwards compatibility will not be provided.


The API accepts GET, POST, PUT, PATCH, and DELETE HTTP requests. For GET requests that accept parameters, required parameters are specified in the path. Optional parameters are accepted as an HTTP query string parameter:


Above, ‘1’, ‘RCP45’, and ‘average_high_temperature’ are required parameters representing :city, :scenario, and :indicator while the optional indicator parameters :years and :units are appended in the query string.

For POST, PUT, PATCH, and DELETE requests, parameters not included in the URL should be encoded as JSON in the request body with a Content-Type of ‘application/json’.


All requests to the Climate API must be authenticated. The Climate API supports the Token authentication scheme via HTTP header.

Your token must be provided in the Authorization HTTP Header with every request. The key should be prefixed by the string literal Token, with whitespace separating the two strings:

Authorization: Token <your token here>

Guard your API token carefully. If compromised, it can be regenerated by going to your Profile. Note that if you regenerate your token, any applications using your old token will cease to operate.

Your token can be found in your Profile while logged in, or by making a POST request to /api-token-auth/.

HTTP Codes

The Climate API uses HTTP codes to indicate a successful, failed, or unsupported request. These codes follow the standards defined in RFC 7231 and more information about any response codes (Including the meaning for codes not mentioned here) can be found there. Below is a list of the most common response codes and what they mean for this API specifically.

200 OK

The request was completed successfully. The response will include a payload with an object containing the requested data based on what API endpoint was requested.

400 Bad Request

The requested endpoint was found, but there was a problem with the query parameters - they may have been in an unexpected format, or asked for something that the API cannot do. The response will include an error message with the invalid parameter and what its expected values are.

404 Not Found

The requested data does not exist - either the URI doesn’t match an API endpoint, or the API couldn’t find the object referenced by a path parameter. If the URI matched a multi-parameter API action, the response will include a message detailing which parameter wasn’t found. Otherwise, check the spelling or consult the API Reference for the path to use for the intended endpoint.

429 Too Many Requests

The rate of user requests exceeded API rate limits - check the error response for details explaining how to proceed. Consult the API Reference for greater detail.

Rate Limiting

In order to ensure fair access to all users of the Climate API, certain endpoints are protected by rate limiting.

If an endpoints is rate limited, the rate limits will be described on that endpoint’s API Reference page.

All endpoints that are rate limited will return HTTP 429 Too Many Requests along with a descriptive message indicating when the endpoint will be available again:

    "detail": "Request was throttled. Expected available in 18.0 seconds."

In addition, the HTTP Header Retry-After will be set with an integer value in seconds that indicates how long to wait until retrying. For example: Retry-After: 43 indicates that the client should retry the same request after waiting at least 43 seconds.

If you have a use case that requires a call volume that exceeds the default rate limits, feel free to contact us.

Cross Origin Resource Sharing (CORS)

As this API is designed for public use, it supports HTTP requests from any origin. To demonstrate, a sample request from an outside domain:

curl -i https://app.climate.azavea.com/api/scenario/ -H "Authorization: Token <your_api_token>" "Origin:https://azavea.com" -X OPTIONS

HTTP/1.1 200 OK
Content-Type: application/json
Date: Thu, 05 Jan 2017 18:45:37 GMT
Server: gunicorn/19.4.5
Vary: Accept
X-Frame-Options: SAMEORIGIN
Content-Length: 398
Connection: keep-alive

How to Make an Indicator Request

Indicators are channels to interact with the Climate API temperature and precipitation data. There are over a dozen highly parametizable indicators that can be used to answer a broad set of research questions.

Maybe you live in or around Los Angeles, California, a megacity surrounded by farms that experiences signficiant drought. As a land use planner, farmer, utility company, mayor, you’re especially interested in preparing for extreme drought. How much drier is it expected to be in LA over the next century?

Get Indicator List

To begin, get the full list of actively developed indicators whose response should should look like:

HTTP/1.1 200 OK
Vary: Accept
Content-Type: application/json

        "name": "total_precipitation",
        "label": "Total Precipitation",
        "description": "Total precipitation",
        "valid_aggregations": [
        "variables": [
        "available_units": [
        "default_units": "in",
        "parameters": [
                "name": "agg",
                "description": "A list of comma separated aggregation types to return. Valid choices are 'min', 'max', 'avg', 'median', 'stddev', 'stdev', and 'XXth'. If using 'XXth', replace the XX with a number between 1-99 to return that percentile. For example, '99th' returns the value of the 99th percentile. The 'XXth' option can be provided multiple times with different values. 'stdev' is an alias to 'stddev'. Defaults to 'min,max,avg'.",
                "required": false,
                "default": "min,max,avg"
                "name": "custom_time_agg",
                "description": "Used in conjunction with the 'custom' time_aggregation value. A list of comma separated month-day pairs defining the time intervals to aggregate within. Data points will only be assigned to one aggregation, and for overlapping intervals the interval defined first will take precedence. Dates are formmatted MM-DD and pairs are formatted 'start:end'. Examples: '3-1:5-31', '1-1:6-30,7-1:12-31'",
                "required": false
                "name": "models",
                "description": "A list of comma separated model names to filter the indicator by. The indicator values in the response will only use the selected models. If not provided, defaults to all models.",
                "required": false
                "name": "time_aggregation",
                "description": "Time granularity to group data by for result structure. Valid aggregations depend on indicator. Can be 'yearly', 'offset_yearly', 'quarterly', 'monthly' or 'custom'. Defaults to 'yearly'. If 'custom', 'custom_time_agg' parameter must be set.",
                "required": false,
                "default": "yearly"
                "name": "units",
                "description": "Units in which to return the data. Defaults to Imperial units (Fahrenheit for temperature indicators and inches for precipitation).",
                "required": false,
                "default": "in"
                "name": "years",
                "description": "A list of comma separated year ranges to filter the response by. Defaults to all years available. A year range is of the form 'start[:end]'. Examples: '2010', '2010:2020', '2010:2020,2030', '2010:2020,2030:2040'",
                "required": false

Each indicator provides a description, parameters, and parameter options. Any special formatting will also be detailed by each param. A closer look at the time_aggregation parameter of total_precipitation reveals it defaults to yearly aggregation, which may be overriden with any of the valid_aggregations or a user-defined date range for custom_time_agg.

Get Indicator Data

To access the indicator data endpoint and retrieve data, we need a city ID, scenario, and indicator. See examples on fetching city and scenario info. The indicator is your choice from the previous indicator list request. Since we’re investigating drought, we’ll look at the total_precipitation indicator, shown above.

Next, let’s use the available parameters to better answer your question. Maybe you’d just like medium and long term annual snapshots, say 2050-2059 and 2090-2099. Maybe you want to prepare for worst and best case scenarios, so you request the minimum and 95th percentile aggregations across models. And maybe you’re Canadian, so we’ll request the data in metric units.

The three paramaters (city id 16, scenario RCP85, and indicator total_precipitation) go directly in the request path:


The remaining parmeters are appended to the above request as a query string, order agnostic, as parameter.name=your_comma_separated_values. For example, the years 2050-2059, 2090-2099 will be years=2050:2059,2090:2099, the aggregation methods agg=min,95th and units units=mm. We do not specify the optional parameters whose default values are acceptable. The final request after appending the query parameters to the request:

GET /api/climate-data/16/RCP85/indicator/total_precipitation/?years=2050:2059,2090:2099&agg=min,95th&units=mm HTTP/1.1
Host: example.org
Authorization: Token 46806a08bf54136e9597e879ed3a0876113fdee6


Note – for brevity, some sections of the mock response have been replaced with ellipses.
HTTP/1.1 200 OK
Vary: Accept
Content-Type: application/json

    "city": {...},
    "scenario": "RCP85",
    "indicator": {...},
    "climate_models": [...],
    "time_aggregation": "yearly",
    "units": "mm",
    "data": {
        "2050": {
            "95th": 502.14717722977383,
            "min": 76.3011513440915
        "2051": {
            "95th": 422.72801741558726,
            "min": 87.7231065219347
        "2052": {
            "95th": 384.35449249837097,
            "min": 69.0049433681225
        "2053": {
            "95th": 400.43015853959912,
            "min": 71.0826842037477
        "2054": {
            "95th": 309.52329147214721,
            "min": 85.7466253202801
        "2055": {
            "95th": 319.07056730291276,
            "min": 43.0824827632023
        "2056": {
            "95th": 355.2163854523539,
            "min": 30.477575477812
        "2057": {
            "95th": 480.81317679456305,
            "min": 78.5377983238959
        "2058": {
            "95th": 398.13347491573677,
            "min": 36.2675766097709
        "2059": {
            "95th": 494.23366352905958,
            "min": 33.8631876410659
        "2092": {
            "95th": 490.50283861029754,
            "min": 66.9922689942837
        "2093": {
            "95th": 476.07359825991239,
            "min": 57.4320739849099
        "2094": {
            "95th": 541.81907089825916,
            "min": 41.4477364907725
        "2095": {
            "95th": 452.86661269688466,
            "min": 100.000476022296
        "2096": {
            "95th": 326.26334963959147,
            "min": 96.7786111861642
        "2091": {
            "95th": 473.53951300371136,
            "min": 29.6483782503742
        "2098": {
            "95th": 489.71526518822668,
            "min": 30.6247628551318
        "2099": {
            "95th": 418.83536519337997,
            "min": 122.807143095861
        "2097": {
            "95th": 397.18454635143212,
            "min": 47.4329927169493
        "2090": {
            "95th": 497.58873546250254,
            "min": 54.6301852183661

Success! To answer our question, we can expect continued, consistent drought mid century and noteworthy turbulence to the norm in precipitation end of century. Notwithstanding, Los Angeles can expect its current drought patterns to be a fairly good measure for future planning. Now that you’ve successfully made this request, consider adjusting the parameters yourself on total_precipitation or query other indicators to get more perspectives on precipitation to further explore drought in Los Angeles.