chatbot works but I don't love it

It's weird to provide examples because of how we preprocessed the data.

Ex: reptile richness is not richness, it's the percent that an area overlaps with the top 20% of reptile rich areas.

This certainly limits the queries we can ask...

app/app.py

@@ -248,10 +248,8 @@ with st.container():
     with st.popover("💬 Example Queries"):
         '''
         Mapping queries:        
-        - Show me areas open to the public that are in the top 10% of species richness.
         - Show me all GAP 1 and 2 lands managed by The Nature Conservancy.
-        - Show me state land smaller than 1000 acres, with a social vulnerability index in the 90th percentile.
-        - Show me GAP 3 and 4 lands managed by BLM in the top 5% of range-size rarity.
+        - Show me GAP 3 and 4 lands with the top 5% of rare amphibian richness.
         - Show me Joshua Tree National Park.
         - Show me all protected lands that have experienced forest fire over at least 50% of their area.
         - Show me the biggest protected area in California. 
@@ -263,7 +261,7 @@ with st.container():
         - What is a GAP code?
         - What percentage of 30x30 conserved land has been impacted by wildfire?
         - What is the total acreage of areas designated as easements?
-        - Who manages the land with the highest amount of irrecoverable carbon and highest social vulnerability index? 
+        - Who manages the land with the highest percentage of wetlands?
         '''
         
         st.info('If the map appears blank, queried data may be too small to see at the default zoom level. Check the table below the map, as query results will also be displayed there.', icon="ℹ️")
@@ -490,25 +488,25 @@ else:
 
 
 # charts displayed based on color_by variable
-amph_chart = bar_chart(df, column, 'mean_amph_richness', "ACE Amphibian Richness")
-reptile_chart = bar_chart(df, column, 'mean_reptile_richness', "ACE Reptile Richness")
-bird_chart = bar_chart(df, column, 'mean_bird_richness', "ACE Bird Richness")
-mammal_chart = bar_chart(df, column, 'mean_mammal_richness', "ACE Mammal Richness")
-rare_amph_chart = bar_chart(df, column, 'mean_rare_amph_richness', "ACE Rare Amphibian Richness")
-rare_reptile_chart = bar_chart(df, column, 'mean_rare_reptile_richness', "ACE Rare Reptile Richness")
-rare_bird_chart = bar_chart(df, column, 'mean_rare_bird_richness', "ACE Rare Bird Richness")
-rare_mammal_chart = bar_chart(df, column, 'mean_rare_mammal_richness', "ACE Rare Mammal Richness")
-end_amph_chart = bar_chart(df, column, 'mean_end_amph_richness', "ACE Endemic Amphibian Richness")
-end_reptile_chart = bar_chart(df, column, 'mean_end_reptile_richness', "ACE Endemic Reptile Richness")
-end_bird_chart = bar_chart(df, column, 'mean_end_bird_richness', "ACE Endemic Bird Richness")
-end_mammal_chart = bar_chart(df, column, 'mean_end_mammal_richness', "ACE Endemic Mammal Richness")
-plant_chart = bar_chart(df, column, 'mean_plant_richness', "Plant Richness")
-rarity_plant_chart = bar_chart(df, column, 'mean_rarityweight_endemic_plant_richness', "Plant Richness")
-wetlands_chart = bar_chart(df, column, 'mean_wetlands', "Wetlands")
-farmland_chart = bar_chart(df, column, 'mean_farmland', "Farmland")
-grazing_chart = bar_chart(df, column, 'mean_grazing', "Grazing land")
-DAC_chart = bar_chart(df, column, 'mean_disadvantaged', "Disadvantaged Communities")
-low_income_chart = bar_chart(df, column, 'mean_low_income', "Low-Income Communities")
+amph_chart = bar_chart(df, column, 'percent_amph_richness', "ACE Amphibian Richness")
+reptile_chart = bar_chart(df, column, 'percent_reptile_richness', "ACE Reptile Richness")
+bird_chart = bar_chart(df, column, 'percent_bird_richness', "ACE Bird Richness")
+mammal_chart = bar_chart(df, column, 'percent_mammal_richness', "ACE Mammal Richness")
+rare_amph_chart = bar_chart(df, column, 'percent_rare_amph_richness', "ACE Rare Amphibian Richness")
+rare_reptile_chart = bar_chart(df, column, 'percent_rare_reptile_richness', "ACE Rare Reptile Richness")
+rare_bird_chart = bar_chart(df, column, 'percent_rare_bird_richness', "ACE Rare Bird Richness")
+rare_mammal_chart = bar_chart(df, column, 'percent_rare_mammal_richness', "ACE Rare Mammal Richness")
+end_amph_chart = bar_chart(df, column, 'percent_end_amph_richness', "ACE Endemic Amphibian Richness")
+end_reptile_chart = bar_chart(df, column, 'percent_end_reptile_richness', "ACE Endemic Reptile Richness")
+end_bird_chart = bar_chart(df, column, 'percent_end_bird_richness', "ACE Endemic Bird Richness")
+end_mammal_chart = bar_chart(df, column, 'percent_end_mammal_richness', "ACE Endemic Mammal Richness")
+plant_chart = bar_chart(df, column, 'percent_plant_richness', "Plant Richness")
+rarity_plant_chart = bar_chart(df, column, 'percent_rarityweight_endemic_plant_richness', "Plant Richness")
+wetlands_chart = bar_chart(df, column, 'percent_wetlands', "Wetlands")
+farmland_chart = bar_chart(df, column, 'percent_farmland', "Farmland")
+grazing_chart = bar_chart(df, column, 'percent_grazing', "Grazing land")
+DAC_chart = bar_chart(df, column, 'percent_disadvantaged', "Disadvantaged Communities")
+low_income_chart = bar_chart(df, column, 'percent_low_income', "Low-Income Communities")
 
 
 

app/system_prompt.txt

@@ -45,16 +45,16 @@ Ensure the response contains only this JSON object, with no additional text, for
 - "acres": Land acreage; measures the size of the area. 
 - "type": Physical type of area, either "Land" or "Water". 
 - "county": County name.
-- "climate_zone": There are 10 climate zones.
+- "climate_zone": 10 climate zones.
 - "habitat_type": California vegetation by wildlife habitat relationship type. There are 13 habitat types: 'Water', 'Conifer Forest', 'Shrub', 'Barren/Other',
        'Hardwood Forest', 'Herbaceous', 'Urban', 'Hardwood Woodland','Desert Shrub', 'Conifer Woodland', 'Wetland', 'Desert Woodland', and 'Agriculture'.
-- "plant_richness": Count of plant species recorded in that area.  
-- "rarityweighted_endemic_plant_richness": Count of rarity-weighted endemic plant species recorded in that area.  
+- "plant_richness": Overlap of top 20% of plant species recorded in that area.  
+- "rarityweighted_endemic_plant_richness": Overlap of top 20% rarity-weighted endemic plant species recorded in that area.  
 - "resilient_connected_network": Categorical variable describing the resilience and connectiveness of that area. 
 There are 12 biodiversity columns of 4 taxonomic groups (amphibian, birds, mammals, reptiles) measured in 3 ways:
-    1) "ACE_amphibian_richness", "ACE_reptile_richness", "ACE_bird_richness", and "ACE_mammal_richness" measure native species richness of overall native diversity  the top 20%; 
-    2) "ACE_rare_amphibian_richness", "ACE_rare_reptile_richness", "ACE_rare_bird_richness", and "ACE_rare_mammal_richness" measure diversity of rare species in the top 5%.
-    3) "ACE_endemic_amphibian_richness", "ACE_endemic_reptile_richness", "ACE_endemic_bird_richness", and "ACE_endemic_mammal_richness" are a weighted measure of endemism that highlights areas that support unique species of limited range. Irreplaceable species in the top 5%. 
+    1) "ACE_amphibian_richness", "ACE_reptile_richness", "ACE_bird_richness", and "ACE_mammal_richness" measure native species richness. These fields represents how much an area overlaps with the top 20% of richness. 
+    2) "ACE_rare_amphibian_richness", "ACE_rare_reptile_richness", "ACE_rare_bird_richness", and "ACE_rare_mammal_richness" measure diversity of rare species. These fields represents how much an area overlaps with the top 5% of rare species. 
+    3) "ACE_endemic_amphibian_richness", "ACE_endemic_reptile_richness", "ACE_endemic_bird_richness", and "ACE_endemic_mammal_richness" are a weighted measure of endemism that highlights areas that support unique species of limited range. These fields represents how much an area overlaps with the top 5% of these irreplaceable species. 
 - "wetlands": How much an area overlaps with wetlands, specifically freshwater emergent wetland, freshwater forested/shrub wetland, or estuarine and marine wetland.
 - "fire": The percentage of the area burned by fires from (2014-2023). Areas can burn more than once, thus the percentage can be above 1
 - "farmland": How much an area overlaps with farmland (specifically prime, unique, of statewide or local importance). 
@@ -89,16 +89,14 @@ example_assistant: {{"sql_query":
 }}
 
 ## Example: 
-example_user: "Who manages the land with the worst biodiversity and highest SVI?"
+example_user: "Who manages the land with the highest rare amphibian richness?"
 example_assistant: {{"sql_query":    
 SELECT "manager", "richness", "svi"
     FROM mydata
     GROUP BY "manager"
-    ORDER BY "richness" ASC, "svi" DESC
+    ORDER BY "ACE_rare_amphibian_richness" DESC
     LIMIT 1;
-"explanation": "I identified the land manager with the worst biodiversity and highest Social Vulnerability Index (SVI) by analyzing the columns: `richness`, which measures species richness, and `svi`, which represents social vulnerability based on factors like socioeconomic status, household characteristics, racial & ethnic minority status, and housing & transportation.
-
-I sorted the data by richness in ascending order (worst biodiversity first) and svi in descending order (highest vulnerability). The result provides the manager, which is the name of the entity managing the land. Note that the manager column refers to the specific agency or organization responsible for managing the land, while `manager_type` categorizes the type of jurisdiction (e.g., Federal, State, Non Profit)."
+"explanation": "I identified the land manager with the highest rare amphibian richness by analyzing the column: `ACE_rare_amphibian_richness`, which has the percent of rare amphibians in the top 5%. I sorted the data in descending order (highest biodiversity first). The result provides the manager, which is the name of the entity managing the land. Note that the manager column refers to the specific agency or organization responsible for managing the land, while `manager_type` categorizes the type of jurisdiction (e.g., Federal, State, Non Profit)."
 }}
 
 ## Example: 
@@ -112,30 +110,26 @@ example_assistant: {{"sql_query":
 "explanation": "I identified the biggest protected area by sorting the data in descending order based on the `acres` column, which represents the size of each area." 
 
 ## Example: 
-example_user: "Show me the 50 most biodiverse areas found in disadvantaged communities."
+example_user: "Show me the 50 most bird biodiverse areas found in disadvantaged communities."
 example_assistant: {{"sql_query":   
-    SELECT "id", "geom", "name", "acres", "richness", "disadvantaged_communities" FROM mydata 
-    WHERE "disadvantaged_communities" > 0
-    ORDER BY "richness" DESC
+    SELECT "id", "geom", "name", "acres", "ACE_bird_richness", "DAC" FROM mydata 
+    WHERE "DAC" > 0
+    ORDER BY "ACE_bird_richness" DESC
     LIMIT 50;
-"explanation": "I used the `richness` column to measure biodiversity and the `disadvantaged_communities` column to identify areas located in disadvantaged communities. The `disadvantaged_communities` value is derived from the Justice40 initiative, which identifies communities burdened by systemic inequities and vulnerabilities across multiple domains, including climate resilience, energy access, health disparities, housing affordability, pollution exposure, transportation infrastructure, water quality, and workforce opportunities.
+"explanation": "I used the `ACE_bird_richness` column to measure bird biodiversity and the `DAC` column to identify areas located in disadvantaged communities. The `DAC` value is derived from the Justice40 initiative, which identifies communities burdened by systemic inequities and vulnerabilities across multiple domains, including climate resilience, energy access, health disparities, housing affordability, pollution exposure, transportation infrastructure, water quality, and workforce opportunities.
 
 The results are sorted in descending order by biodiversity richness (highest biodiversity first), and only areas with a `disadvantaged_communities` value greater than 0 (indicating some portion of the area overlaps with a disadvantaged community) are included."
 }}
 
 ## Example: 
-example_user: "Show me federally managed gap 3 lands that are in the top 5% of biodiversity richness and have experienced forest fire over at least 50% of their area"
+example_user: "Show me federally managed gap 3 lands that are in the top 5% of endemic mammal richness and have experienced forest fire over at least 50% of their area"
 sql_query:  
-    WITH temp AS (
-        SELECT PERCENTILE_CONT(0.95) WITHIN GROUP (ORDER BY "richness") AS richness_95_percentile
-        FROM mydata
-    )
-    SELECT "id", "geom", "name", "acres", "richness", "gap_code", "fire"
+    SELECT "id", "geom", "name", "acres", "ACE_endemic_mammal_richness", "gap_code", "fire"
     FROM mydata
     WHERE "gap_code" = 3
         AND "fire" >= 0.5
-        and "manager_type" = 'Federal'
-        AND "richness" > (SELECT richness_95_percentile FROM temp);
+        AND "manager_type" = 'Federal'
+        AND "ACE_endemic_mammal_richness" > 0;
 
 ## Example: 
 example_user: "What is the total acreage of areas designated as easements?
@@ -145,23 +139,23 @@ sql_query:
     WHERE "easement" = 'True';
 
 ## Example: 
-example_user: "Which ecoregions are in the top 10% of range-size rarity?"
+example_user: "Which ecoregions are in the top 50% of farmland?"
 sql_query:  
     WITH temp AS (
-        SELECT PERCENTILE_CONT(0.90) WITHIN GROUP (ORDER BY "rsr") AS rsr_90_percentile
+        SELECT PERCENTILE_CONT(0.50) WITHIN GROUP (ORDER BY "farmland") AS farmland_50_percentile
         FROM mydata
     )
     SELECT "ecoregion"
     FROM mydata
-    WHERE "rsr" > (SELECT rsr_90_percentile FROM temp);
+    WHERE "farmland" > (SELECT farmland_90_percentile FROM temp);
 
 ## Example: 
-example_user: "Show me protected lands in disadvantaged communities that have had prescribed fires in at least 30% of its area."
+example_user: "Show me protected lands in disadvantaged communities that have wetlands in at least 30% of its area."
 sql_query:  
-    SELECT "id", "geom", "name", "acres", "rxburn", "percent_disadvantaged" 
+    SELECT "id", "geom", "name", "acres", "wetlands", "DAC" 
     FROM mydata 
-    WHERE "percent_disadvantaged" > 0
-    AND "percent_rxburn_10yr" >= 0.3;
+    WHERE "DAC" > 0
+    AND "wetlands" >= 0.3;
 
 
 

app/utils.py

@@ -95,26 +95,26 @@ def get_summary(ca, combined_filter, column, main_group, colors = None):
     df = (df.group_by(*column)
           .aggregate(percent_CA=(_.acres.sum() / ca_area_acres),
                      acres=_.acres.sum(),
-                     mean_amph_richness =(_.ACE_amphibian_richness * _.acres).sum() / _.acres.sum(),
-                     mean_reptile_richness=(_.ACE_reptile_richness * _.acres).sum() / _.acres.sum(),
-                     mean_bird_richness=(_.ACE_bird_richness * _.acres).sum() / _.acres.sum(),
-                     mean_mammal_richness=(_.ACE_mammal_richness * _.acres).sum() / _.acres.sum(),
-                     mean_rare_amph_richness =(_.ACE_rare_amphibian_richness * _.acres).sum() / _.acres.sum(),
-                     mean_rare_reptile_richness=(_.ACE_rare_reptile_richness * _.acres).sum() / _.acres.sum(),
-                     mean_rare_bird_richness=(_.ACE_rare_bird_richness * _.acres).sum() / _.acres.sum(),
-                     mean_rare_mammal_richness=(_.ACE_rare_mammal_richness * _.acres).sum() / _.acres.sum(),
-                     mean_end_amph_richness =(_.ACE_endemic_amphibian_richness * _.acres).sum() / _.acres.sum(),
-                     mean_end_reptile_richness=(_.ACE_endemic_reptile_richness * _.acres).sum() / _.acres.sum(),
-                     mean_end_bird_richness=(_.ACE_endemic_bird_richness * _.acres).sum() / _.acres.sum(),
-                     mean_end_mammal_richness=(_.ACE_endemic_mammal_richness * _.acres).sum() / _.acres.sum(),
-                     mean_plant_richness=(_.plant_richness * _.acres).sum()/_.acres.sum(),
-                     mean_rarityweight_endemic_plant_richness=(_.rarityweighted_endemic_plant_richness * _.acres).sum()/_.acres.sum(),
-                     mean_wetlands=(_.wetlands * _.acres).sum()/_.acres.sum(),
-                     mean_fire=(_.fire * _.acres).sum() / _.acres.sum(),
-                     mean_farmland=(_.farmland * _.acres).sum() / _.acres.sum(),
-                     mean_grazing=(_.grazing * _.acres).sum() / _.acres.sum(),
-                     mean_disadvantaged=(_.DAC * _.acres).sum() / _.acres.sum(),
-                     mean_low_income=(_.low_income * _.acres).sum() / _.acres.sum()
+                     percent_amph_richness =(_.ACE_amphibian_richness * _.acres).sum() / _.acres.sum(),
+                     percent_reptile_richness=(_.ACE_reptile_richness * _.acres).sum() / _.acres.sum(),
+                     percent_bird_richness=(_.ACE_bird_richness * _.acres).sum() / _.acres.sum(),
+                     percent_mammal_richness=(_.ACE_mammal_richness * _.acres).sum() / _.acres.sum(),
+                     percent_rare_amph_richness =(_.ACE_rare_amphibian_richness * _.acres).sum() / _.acres.sum(),
+                     percent_rare_reptile_richness=(_.ACE_rare_reptile_richness * _.acres).sum() / _.acres.sum(),
+                     percent_rare_bird_richness=(_.ACE_rare_bird_richness * _.acres).sum() / _.acres.sum(),
+                     percent_rare_mammal_richness=(_.ACE_rare_mammal_richness * _.acres).sum() / _.acres.sum(),
+                     percent_end_amph_richness =(_.ACE_endemic_amphibian_richness * _.acres).sum() / _.acres.sum(),
+                     percent_end_reptile_richness=(_.ACE_endemic_reptile_richness * _.acres).sum() / _.acres.sum(),
+                     percent_end_bird_richness=(_.ACE_endemic_bird_richness * _.acres).sum() / _.acres.sum(),
+                     percent_end_mammal_richness=(_.ACE_endemic_mammal_richness * _.acres).sum() / _.acres.sum(),
+                     percent_plant_richness=(_.plant_richness * _.acres).sum()/_.acres.sum(),
+                     percent_rarityweight_endemic_plant_richness=(_.rarityweighted_endemic_plant_richness * _.acres).sum()/_.acres.sum(),
+                     percent_wetlands=(_.wetlands * _.acres).sum()/_.acres.sum(),
+                     percent_fire=(_.fire * _.acres).sum() / _.acres.sum(),
+                     percent_farmland=(_.farmland * _.acres).sum() / _.acres.sum(),
+                     percent_grazing=(_.grazing * _.acres).sum() / _.acres.sum(),
+                     percent_disadvantaged=(_.DAC * _.acres).sum() / _.acres.sum(),
+                     percent_low_income=(_.low_income * _.acres).sum() / _.acres.sum()
                     )
           .mutate(percent_CA=_.percent_CA.round(5), acres=_.acres.round(0))
          )
@@ -324,7 +324,7 @@ def get_chart_settings(x, stacked):
         "mean_svi": "SVI (Mean)", "mean_fire": "Fire (Mean)", "mean_rxburn": "Rx Fire (Mean)"
     }
 
-    angle = 270 if x in ["manager_type", "ecoregion"] else 0
+    angle = 270 if x in ["manager_type", "ecoregion", "status", "habitat_type", "resilient_connected_network"] else 0
     height = 250 if stacked else 400 if x == "ecoregion" else 350 if x == "manager_type" else 300
 
     return sort_options.get(x, "x"), angle, height, y_titles.get(x, x)