Merge pull request #261 from lxiam26/main

Documentation Updates

Author: dwreeves

app/static/images/boathouse_map.png

@@ -329,6 +329,15 @@ nav a:hover {
   z-index: 0;
 }
 
+.map_img {
+  display: block;
+  margin: 20px auto;
+  max-width: 645px;
+  width: 100%;
+  height: auto;
+}
+
+
 .auth-container {
   display: flex;
   flex-direction: row;

app/static/style.css

@@ -3,7 +3,7 @@
 {% block content %}
     <h2>Flagging Program</h2>
     <p>
-        Of the many services that the CRWA provides to the greater Boston community, one of those is monitoring whether it is safe to boat in the Charles River.The CRWA
+        Of the many services that the CRWA provides to the greater Boston community, one of those is monitoring whether it is safe to boat in the Charles River. The CRWA
         Flagging Program uses a system of color-coded flags to indicate whether or not the river's water quality is safe for boating at 13 boating locations between Watertown and Boston.
         Flag colors are based on <i>E. coli</i> and cyanobacteria (blue-green algae) levels, as well as if a CSO discharge has occurred in the last 48 hours.
         Blue flags indicate suitable boating conditions and red flags indicate potential health risks.
@@ -17,10 +17,14 @@ <h2>Flagging Program</h2>
             </iframe>
         </div>
     </div>
-    <h2>Water Contamination</h2>
+    <h2>Water Quality</h2>
+    <p>
+        CRWA conducts water quality sampling every week during the peak recreational season in the Lower Basin
+        (final 10 miles of the river from the Watertown Dam to the New Charles River Dam) from May to late October.
+        During monitoring events, CRWA measures water temperature and depth and collects water quality samples at four sampling locations.
+    </p>
+    <img class="map_img" src="{{ url_for('static', filename='images/boathouse_map.png') }}" alt="Map of the 13 boathouses for the 4 reaches (as of 2025)"/>
     <p>
-        CRWA conducts water quality sampling every week during the peak recreational season in the Lower Basin from May to late October.
-        During monitoring events, CRWA measures water temperature and depth and collects water quality samples at four sampling locations in the Lower Basin.
         G&L Laboratory in Quincy analyzes the samples for <i>E. coli</i> bacteria. Water temperature and depth are measured in situ with a digital field thermometer
         and a digital depth finder, respectively. Sampling locations are center channel sites, upstream of the following bridges: North Beacon Street,
         Larz Anderson Memorial Bridge, Boston University Bridge, and Longfellow Bridge. All samples are manual grab samples collected from a boat.
@@ -43,20 +47,20 @@ <h3>Cyanobacteria</h3>
         Cyanobacteria, also called blue-green algae, are microscopic cells that naturally grow in rivers and lakes. Cyanobacteria populations can explode in warm water with excess nutrients, and exposure
         to these blooms can have negative health effects for humans, animals, and ecosystems. CRWA monitors cyanobacteria levels in the Charles River Basin in Boston and Cambridge during the summer and
         reports high levels to the Department of Public Health.
-        (<a href="https://www.crwa.org/monitor.html">CRWA Website</a>)
+        (<a href="https://www.crwa.org/cyano-monitoring">CRWA Website</a>)
     </p>
     <h2>Predictive Models</h2>
     <p>
         As part of our public notification program, CRWA uses statistical models to predict the likelihood that water quality will be in violation of the State's recreational water quality standard threshold value at each of our four sampling reaches.
-        The models use antecedent rainfall, river flow, wind speed and air and water temperature to make predictions every hour throughout the recreational season. Models are necessary to produce real time water quality forecasts
+        The models use antecedent rainfall, river flow, relative humidity, photosynthetic active radiation, gage height, dew point, and pressure to make predictions every hour throughout the recreational season. Models are necessary to produce real time water quality forecasts
         because <i>E. coli</i> analysis requires at least 24 hours between sample collection and result reporting. These hourly forecasts allow boaters to make informed decisions about their desired contact with the river based on current
         river water quality conditions.
         (<a href="https://www.crwa.org/s/2024-Flagging-Report_FINAL-1.pdf">2024 Flagging Report</a>)
     </p>
     <p>
         CRWA's predictive models utilize live data from a HOBOlink RX3000 Remote Monitoring Station device, which is
         stationed at a boathouse on the Charles River. (You can download the latest raw data from this device
-        <a href="https://www.hobolink.com/p/0cdac4a6910cef5a8883deb005d73ae1">here</a>.) Additionally, the model utilizes
+        <a href="Update with new link https://hobolink.com/p/279ab82333aa3378b932370ed5ddc712">here</a>.) Additionally, the model utilizes
         data from two USGS stream gauges located in Waltham and Brookline.</a>
     </p>
     <h2>Credits</h2>
@@ -69,11 +73,9 @@ <h2>Credits</h2>
         <p>
             The predictive model currently being used for the flagging program website was developed and deployed by <a href="https://www.linkedin.com/in/laura_xia" target="_blank">Laura Xia</a>
             (MIT, class of 2026).
-             <!-- She has updated the model with new data to reflect the changing conditions of the Charles River. -->
-
-            <!-- We can also write a quick sentence about the model and how it was developed if you'd like.
-            "In addition to updating the model with new data to reflect the changing conditions of the Charles River, Laura's model improves on previous modeling efforts by replacing the older model's logistic regression with a regression on log(E. coli cfu/100ml), using machine learning methods for feature selection, and using [hierarchical modeling? partial pooling? mixed effects?] to regularize inter-reach model parameters." Jumping the gun on the last one though. 😂 You should phrase the description of the model yourself, though, I'm only offering this as a starting point. Also, we'll obviously wait on a description until you think it's done. -->
-
+            She has updated the model with new data to reflect the changing conditions of the Charles River.
+            She has also updated the model itself to regress on the log E. coli concentration directly instead of using a logistic regression.
+            Various aspects like feature selection, weight balancing, and mixed effects were explored in depth in this process.
         </p>
         <p>
             This website is open source. You can view the source code and a list of contributors

app/templates/about.html

@@ -34,7 +34,7 @@ <h2>About CRWA's Water Quality Models</h2>
     who recommends bloom advisories.</li>
     <li>If <strong>CRWA's water quality models</strong> predict <i>E. coli</i> bacteria levels that exceed the State's
     recreational water quality standard threshold value, red flags will be flown at the affected boathouses. CRWA estimates <i>E. coli</i>
-    bacteria levels using models that take into account a variety of environmental factors including rainfall and river
+    bacteria levels using models that take into account rainfall and river
     flow. CRWA collects water samples weekly to verify the model forecasts.</li>
 </ol>
 <h2>Stay Updated</h2>