Merge pull request #12 from limnoliver/cleanup

Metadata cleanup

2_observations.yml

@@ -12,7 +12,6 @@ targets:
   2_observations:
     depends:
       - out_data/temperature_observations.csv
-      - out_data/flow_observations.csv
 
   # daily flow and temperature data
   out_data/temperature_observations.csv:
@@ -21,8 +20,3 @@ targets:
       var = I('wtemp(C)'),
       in_file = 'in_data/Data - ERL paper/Forcing_attrFiles/no_dam_forcing_60__days118sites.csv')
 
-  out_data/flow_observations.csv:
-    command: extract_obs(
-      out_file = target_name,
-      var = I('discharge(cfs)'),
-      in_file = 'in_data/Data - ERL paper/Forcing_attrFiles/no_dam_forcing_60__days118sites.csv')

3_inputs.yml

@@ -17,6 +17,7 @@ targets:
       - out_data/AT_basin_attributes.csv
       - out_data/weather_drivers.zip
       - out_data/pred_discharge.csv
+      - out_data/obs_discharge.csv
 
   out_data/AT_basin_attributes.csv:
     command: extract_AT_attributes(
@@ -29,8 +30,14 @@ targets:
   out_data/weather_drivers.zip:
     command: zip_this(out_file = target_name, weather_drivers)
 
-
   out_data/pred_discharge.csv:
     command: subset_pred_discharge(
       out_file = target_name,
       in_file = 'in_data/Data - ERL paper/Forcing_attrFiles/no_dam_forcing_60__days118sites.csv')
+
+  out_data/obs_discharge.csv:
+    command: extract_obs(
+      out_file = target_name,
+      var = I('discharge(cfs)'),
+      in_file = 'in_data/Data - ERL paper/Forcing_attrFiles/no_dam_forcing_60__days118sites.csv')
+

in_text/text_01_spatial.yml

@@ -38,23 +38,27 @@ entities:
       attr-def: >-
         Latitude of the site location.
       attr-defs: NA
-      data-min: NA
-      data-max: NA
-      data-units: NA
+      data-min: 30.14549
+      data-max: 48.90596
+      data-units: decimal degrees
     -
       attr-label: long
       attr-def: >-
         Longitude of the site location.
       attr-defs: NA
-      data-min: NA
-      data-max: NA
-      data-units: NA
+      data-min: -123.3299
+      data-max: -70.97964
+      data-units: decimal degrees
 
 
 data-name: GIS points of sites used in this study.
 data-description: Location of USGS river gages used in this study.
 
 file-format: Shapefile Data Set
+process-date: 20201028
+indirect-spatial: U.S.A.
+latitude-res: 0.00001
+longitude-res: 0.00001
 
 build-environment: >-
   This dataset was generated using XX.

in_text/text_02_observations.yml

@@ -3,7 +3,7 @@ title: >-
 
 
 abstract: >-
-  Mean daily temperature and discharge observations retrieved from NWIS. The temperature observations were used to train and validate all temperature models, while flow observations were used as a model input during training for a subset of the temperature models. The model training period was from 2010-10-01 to 2014-09-30, and the test period was from 2014-10-01 to 2016-09-30.
+  Mean daily temperature and discharge observations retrieved from NWIS. The temperature observations were used to train and validate all temperature models. The model training period was from 2010-10-01 to 2014-09-30, and the test period was from 2014-10-01 to 2016-09-30.
 
 cross-cites:
   -
@@ -41,32 +41,13 @@ entities:
       data-min: NA
       data-max: NA
       data-units: degrees Celsius
-  -
-    data-name: flow_observations.csv
-    data-description: Observed mean daily discharge observation retrieved from NWIS for the 118 gages used in this study. Flow observations were used as a driver in the water temperature model. The data were retrieved from NWIS and are limited to the test and training period, from 2010-10-01 through 2016-09-30.
-    attributes:
-    -
-      attr-label: site_no
-      attr-def: >-
-        USGS unique site identifier.
-      attr-defs: NA
-      data-min: NA
-      data-max: NA
-      data-units: NA
-    -
-      attr-label: datetime
-      attr-def: >-
-        Date of temperature observation.
-      attr-defs: NA
-      data-min: NA
-      data-max: NA
-      data-units: NA
-    -
-      attr-label: discharge(cfs)
-      attr-def: Observed mean daily discharge
-      attr-defs: NA
-      data-min: NA
-      data-max: NA
-      data-units: degrees Celsius
 
-file-format: comma-delimited files
+
+process-date: XX
+indirect-spatial: U.S.A.
+latitude-res: 0.00001
+longitude-res: 0.00001
+data-name: Water temperature observations
+data-description: >-
+  Water temperature observations used to train and validate models described in Rahmani et al. 2020.
+file-format: comma seperated file format (csv)

in_text/text_03_inputs.yml

@@ -2,7 +2,7 @@ title: >-
   Exploring the exceptional performance of a deep learning stream temperature model and the value of streamflow data: 3 model inputs
 
 abstract: >-
-  Inputs to the deep learning models included daily weather forcing data, as well as river catchment attributes.
+  Inputs to the deep learning models included daily weather forcing data, river catchment attributes, and simulated or observed flow.
 
 cross-cites:
   -
@@ -305,14 +305,41 @@ entities:
       data-min: NA
       data-max: NA
       data-units: cubic feet per second
-
-
+  -
+    data-name: flow_observations.csv
+    data-description: Observed mean daily discharge observation retrieved from NWIS for the 118 gages used in this study. Flow observations were used as a driver in the water temperature model. The data were retrieved from NWIS and are limited to the test and training period, from 2010-10-01 through 2016-09-30.
+    attributes:
+    -
+      attr-label: site_no
+      attr-def: >-
+        USGS unique site identifier.
+      attr-defs: NA
+      data-min: NA
+      data-max: NA
+      data-units: NA
+    -
+      attr-label: datetime
+      attr-def: >-
+        Date of temperature observation.
+      attr-defs: NA
+      data-min: NA
+      data-max: NA
+      data-units: NA
+    -
+      attr-label: discharge(cfs)
+      attr-def: Observed mean daily discharge
+      attr-defs: NA
+      data-min: NA
+      data-max: NA
+      data-units: degrees Celsius
 
 build-environment: Multiple computer systems were used to generate these data, including XX. The open source languages R and Python was used on all systems, as well as XX.
 
 process-date: !expr format(Sys.time(),'%Y%m%d')
 indirect-spatial: U.S.A.
 latitude-res: 0.1
 longitude-res: 0.1
-data-name: weather data, river catchment metadata
-
+data-name: Model driver data
+data-description: >-
+  Inputs (drivers) for the temperature models described in Rahmani et al. 2020, including weather drivers, river basin attributes, and simulated and observed river discharge.
+file-format: comma seperated file format (csv)

in_text/text_05_predictions.yml

@@ -2,15 +2,15 @@ title: >-
     Exploring the exceptional performance of a deep learning stream temperature model and the value of streamflow data: 5 model prediction data
 
 abstract: >-
-  A deep learning model framework was used to make water temperature predictions in 118 river catchments across the U.S. All four model (LR, noQ, obsQ, and simQ) predictions are included. Additionally, a deep learning model was used to simulate discharge, which was used as inputs to the water temperature model.
+  A deep learning model framework was used to make water temperature predictions in 118 river catchments across the U.S. All four model (LR, noQ, obsQ, and simQ) predictions are included.
 
 cross-cites:
   -
-    authors: ['XX']
+    authors: ['DP Feng', 'K Fang', "CP Shen"]
     title: >-
-      Cross cite code base?
-    pubdate: XX
-    link: XX
+      Enhancing streamflow forecast and extracting insights using continental-scale long-short term memory networks with data integration
+    pubdate: 2020
+    link: https://doi.org/10.1029/2019WR026793
 
 build-environment: >-
   We used XX open source XX; Any supercomputing resources used? XX
@@ -46,3 +46,12 @@ entities:
       data-min: NA
       data-max: NA
       data-units: degrees Celsius
+
+process-date: !expr format(Sys.time(),'%Y%m%d')
+indirect-spatial: U.S.A.
+latitude-res: 0.00001
+longitude-res: 0.00001
+data-name: Model predictions
+data-description: >-
+  Stream water temperature predictions from each model described in Rahmani et al. 2020.
+file-format: comma seperated file format (csv)

in_text/text_06_evaluation.yml

@@ -77,3 +77,12 @@ entities:
 
 build-environment: >-
   We used XX open source XX.
+
+process-date: !expr format(Sys.time(),'%Y%m%d')
+indirect-spatial: U.S.A.
+latitude-res: 0.00001
+longitude-res: 0.00001
+data-name: Model evaluation metrics
+data-description: >-
+  Evaluation metrics used to compare stream temperature models in Rahmani et al. 2020.
+file-format: comma seperated file format (csv)

in_text/text_SHARED.yml

@@ -53,7 +53,13 @@ funding-credits: >-
 process-description: >-
   At the core of the modeling framework is a deep learning model that uses inputs of XX.
 
+process-date: 20201028
+latitude-res: 0.1
+longitude-res: 0.1
 distro-person: Samantha K. Oliver
+
+build-environment: Multiple computer systems were used to generate these data, including linux, OSX. The open source languages R and Python were used on all systems. XX
+
 liability-statement: >-
   Unless otherwise stated, all data, metadata and related materials are considered to satisfy the quality standards relative to the purpose for which the data were collected.
   Although these data and associated metadata have been reviewed for accuracy and completeness and approved for release by the U.S. Geological Survey (USGS),