jacob-a-brown · May 19, 2026 · Apr 27, 2026 · Apr 27, 2026 · Apr 27, 2026 · Apr 28, 2026
diff --git a/README.md b/README.md
@@ -38,8 +38,9 @@ Data comes from the following sources. We are continuously adding new sources as
   - Available data: `water levels`
 - [Pecos Valley Artesian Conservancy District (PVACD)](https://st2.newmexicowaterdata.org/FROST-Server/v1.1/Locations?$filter=properties/agency%20eq%20%27PVACD%27)
   - Available data: `water levels`
-- [USGS (NWIS)](https://waterdata.usgs.gov/nwis)
+- [USGS (NWIS)](https://api.waterdata.usgs.gov/docs/)
   - Available data: `water levels`
+  - **IMPORTANT:** The USGS now uses API keys. To prevent yourself from hitting the rate limit please [acquire an API key](https://api.waterdata.usgs.gov/signup/), save it, and provide it via the `--usgs-api-key` flag when gathering water level or site data from the USGS.
 - [Water Quality Portal (WQP)](https://www.waterqualitydata.us/)
   - Available data: `water levels`, `water quality`
 
@@ -189,6 +190,20 @@ The Data Integration Engine enables the user to obtain groundwater level and gro
 - `--no-pvacd` to exclude Pecos Valley Artesian Convservancy District (PVACD) data
 - `--no-wqp` to exclude Water Quality Portal (WQP) data
 
+### USGS API Keys
+
+The USGS now uses [API keys](https://api.waterdata.usgs.gov/signup/) to increase the query rate limit for their APIs. If you intend to include USGS water level data in your output please acquire an API key, save it somewhere, and provide it via the `--usgs-api-key` flag. For example:
+
+```
+die weave waterlevels --output-type timeseries_unified --usgs-api-key FAKE_API_KEY
+```
+
+or
+
+```
+die sites --usgs-api-key FAKE_API_KEY
+```
+
 ### Geographic Filters [In Development]
 
 The following flags can be used to geographically filter data:

diff --git a/backend/config.py b/backend/config.py
@@ -281,8 +281,19 @@ def get_config_and_false_agencies(self):
                 "nwis",
                 "pvacd",
             ]
+        elif self.parameter in [BICARBONATE]:
+            config_agencies = ["nmbgmr_amp", "nmed_dwb", "nmose_isc_seven_rivers", "wqp"]
+            false_agencies = [
+                "bor",
+                "bernco",
+                "cabq",
+                "ebid",
+                "nmose_roswell",
+                "nmose_pod",
+                "nwis",
+                "pvacd",
+            ]
         elif self.parameter in [
-            BICARBONATE,
             CALCIUM,
             CHLORIDE,
             FLUORIDE,

diff --git a/backend/connectors/bor/source.py b/backend/connectors/bor/source.py
@@ -48,16 +48,16 @@ def __repr__(self):
 
     def health(self):
         try:
-            self.get_records()
-            return True
+            resp = self.get_records()
+            return bool(resp)
         except Exception:
             return False
 
     def get_records(self):
         # locationTypeId 10 is for wells
         url = "https://data.usbr.gov/rise/api/location"
         params = {"stateId": "NM", "locationTypeId": 10}
-        return self._execute_json_request(url, params)
+        return self._execute_json_request(url, params, tag="data")
 
 
 def parse_dt(dt):
@@ -119,13 +119,14 @@ def get_records(self, site_record):
         code = get_analyte_search_param(self.config.parameter, BOR_ANALYTE_MAPPING)
 
         catalog_record_data = self._execute_json_request(
-            f"https://data.usbr.gov{site_record.catalogRecords[0]['id']}"
+            f"https://data.usbr.gov{site_record.catalogRecords[0]['id']}",
+            tag="data"
         )
         catalog_items = catalog_record_data["relationships"]["catalogItems"]["data"]
 
         for i, item in enumerate(self._reorder_catalog_items(catalog_items)):
 
-            data = self._execute_json_request(f'https://data.usbr.gov{item["id"]}')
+            data = self._execute_json_request(f'https://data.usbr.gov{item["id"]}', tag="data")
             if not data:
                 continue
 
@@ -142,6 +143,7 @@ def get_records(self, site_record):
                 return self._execute_json_request(
                     "https://data.usbr.gov/rise/api/result",
                     params={"itemId": data["attributes"]["_id"]},
+                    tag="data"
                 )
 
 

diff --git a/backend/connectors/isc_seven_rivers/source.py b/backend/connectors/isc_seven_rivers/source.py
@@ -76,15 +76,16 @@ def __repr__(self):
 
     def health(self):
         try:
-            self.get_records()
-            return True
+            resp = self.get_records()
+            return bool(resp)
         except Exception as e:
             print("Failed to get records", e)
             return False
 
     def get_records(self):
         return self._execute_json_request(
             _make_url("getMonitoringPoints.ashx"),
+            tag="data",
         )
 
 
@@ -100,7 +101,7 @@ def _get_analyte_id_and_name(self, analyte):
         """ """
         if self._analyte_ids is None:
 
-            resp = self._execute_json_request(_make_url("getAnalytes.ashx"))
+            resp = self._execute_json_request(_make_url("getAnalytes.ashx"), tag="data")
             if resp:
                 self._analyte_ids = {r["name"]: r["id"] for r in resp}
 
@@ -164,7 +165,7 @@ def get_records(self, site_record):
                 self._source_parameter_name = analyte_id_and_name["name"]
 
             return self._execute_json_request(
-                _make_url("getReadings.ashx"), params=params
+                _make_url("getReadings.ashx"), params=params, tag="data"
             )
 
 
@@ -184,6 +185,7 @@ def get_records(self, site_record):
         return self._execute_json_request(
             _make_url("getWaterLevels.ashx"),
             params=params,
+            tag="data",
         )
 
     def _clean_records(self, records):

diff --git a/backend/connectors/nmbgmr/source.py b/backend/connectors/nmbgmr/source.py
@@ -45,6 +45,12 @@
 )
 
 
+# Set timeout to 15 minutes for analyte and water level requests since some sites have a large number of records and the NMBGMR API can be slow to respond.
+# Don't use timeout=None since that can cause the request to hang indefinitely if there are issues with the API.
+# Instead, catch timeout and other exceptions and retry the request up to 7 times with a delay between retries.
+
+TIMEOUT=15*60
+
 def _make_url(endpoint):
     if os.getenv("DEBUG") == "1":
         url = f"http://localhost:8000/latest/{endpoint}"
@@ -62,10 +68,13 @@ def __repr__(self):
         return "NMBGMRSiteSource"
 
     def health(self):
-        resp = self._execute_json_request(
-            _make_url("locations"), tag="features", params={"limit": 1}
-        )
-        return bool(resp)
+        try:
+            resp = self._execute_json_request(
+                _make_url("locations"), tag="features", params={"limit": 1}
+            )
+            return bool(resp)
+        except Exception:
+            return False
 
     def get_records(self):
         config = self.config
@@ -82,10 +91,10 @@ def get_records(self):
             else:
                 params["parameter"] = "Manual groundwater levels"
 
-        # tags="features" because the response object is a GeoJSON
         sites = self._execute_json_request(
-            _make_url("locations"), params, tag="features", timeout=30
+            _make_url("locations"), params, tag="features", timeout=TIMEOUT
         )
+
         if not config.sites_only:
             for site in sites:
                 if get_bool_env_variable("IS_TESTING_ENV"):
@@ -119,14 +128,17 @@ def get_records(self, site_record):
         analyte = get_analyte_search_param(
             self.config.parameter, NMBGMR_ANALYTE_MAPPING
         )
+
         records = self._execute_json_request(
             _make_url("waterchemistry"),
             params={
                 "pointid": ",".join(make_site_list(site_record)),
                 "analyte": analyte,
             },
             tag="",
+            timeout=TIMEOUT
         )
+
         records_sorted_by_pointid = {}
         for pointid in records.keys():
             records_sorted_by_pointid[pointid] = records[pointid][analyte]
@@ -224,15 +236,18 @@ def get_records(self, site_record):
         # just use manual waterlevels temporarily
         url = _make_url("waterlevels/manual")
 
-        paginated_records = self._execute_json_request(url, params, tag="")
+        paginated_records = self._execute_json_request(url, params, tag="", timeout=TIMEOUT)
+
         items = paginated_records["items"]
         page = paginated_records["page"]
         pages = paginated_records["pages"]
 
         while page < pages:
             page += 1
             params["page"] = page
-            new_records = self._execute_json_request(url, params, tag="")
+
+            new_records = self._execute_json_request(url, params, tag="", timeout=TIMEOUT)
+
             items.extend(new_records["items"])
             pages = new_records["pages"]
 

diff --git a/backend/connectors/nmenv/source.py b/backend/connectors/nmenv/source.py
@@ -45,8 +45,12 @@ def __repr__(self):
         return "DWBSiteSource"
 
     def health(self):
-        return self.get_records(top=10, analyte=TDS)
-
+        try:
+            resp = self.get_records(top=10, analyte=TDS)
+            return bool(resp)
+        except Exception:
+            return False
+
     def get_records(self, *args, **kw):
 
         analyte = None

diff --git a/backend/connectors/st_connector.py b/backend/connectors/st_connector.py
@@ -90,8 +90,14 @@ def make_dt_filter(tag, start, end):
 
 class STSiteSource(BaseSiteSource, STSource):
     def health(self):
-        return self.get_records(top=10)
-
+        try:
+            # Health checks may run before config is initialized; probe the service directly.
+            service = self.get_service()
+            resp = list(service.locations().query().top(1).list())
+            return bool(resp)
+        except Exception:
+            return False
+
     def get_records(self, *args, **kw):
         service = self.get_service()