Spaces:

LunaAmagi
/

chronostasis

Sleeping

App Files Files Community

LunaAmagi commited on Apr 26

Commit

272d4d3

1 Parent(s): ea1d3c5

feat: 15 regions seasonal data

Browse files

Files changed (1) hide show

tasks.py +567 -256

tasks.py CHANGED Viewed

@@ -1,319 +1,630 @@
 """
-tasks.py — Region-aware OpenEnv Tasks for Chronostasis
 """
 import re
-from abc import ABC, abstractmethod
 from typing import Any, Dict, List, Optional
-def _safe_float(s):
-    try:
-        return float(str(s).replace(",", "").strip())
-    except Exception:
-        return None
-def _extract_nums(text):
-    return [x for x in [_safe_float(n) for n in re.findall(r"\d[\d,.]*", text)] if x is not None]
-REGIONS = {
     "brahmaputra": {
-        "name": "Brahmaputra Valley", "state": "Assam", "river": "Brahmaputra",
-        "flood_areas": {2022: 4812.3, 2023: 3601.7, 2024: 4101.2},
-        "peak_year": 2022, "chronic_km2": 1247.6, "chronic_pop": 2_400_000,
-        "chronic_districts": ["Morigaon", "Dhubri", "Barpeta", "Goalpara", "Kamrup"],
-        "high_risk_zones": ["lower Brahmaputra floodplain", "Dhubri district riverbank", "Barpeta wetland belt", "Morigaon char lands"],
-        "accuracy_pct": 92.39, "risk_zones_km2": {"high": 3218.4, "moderate": 5901.2, "low": 8240.1},
-        "peak_rainfall_mm": 1500, "sar_threshold_db": -16,
     },
     "ganga": {
-        "name": "Ganga Plains", "state": "Bihar", "river": "Ganga",
-        "flood_areas": {2022: 6241.8, 2023: 4987.3, 2024: 5614.6},
-        "peak_year": 2022, "chronic_km2": 2108.4, "chronic_pop": 3_800_000,
-        "chronic_districts": ["Darbhanga", "Sitamarhi", "Madhubani", "Saharsa", "Supaul"],
-        "high_risk_zones": ["North Bihar Kosi belt", "Darbhanga low-lying plains", "Gandak floodplain", "Bagmati river corridor"],
-        "accuracy_pct": 89.74, "risk_zones_km2": {"high": 4812.1, "moderate": 7234.5, "low": 9801.2},
-        "peak_rainfall_mm": 1200, "sar_threshold_db": -16,
     },
     "mahanadi": {
-        "name": "Mahanadi Delta", "state": "Odisha", "river": "Mahanadi",
-        "flood_areas": {2022: 3142.7, 2023: 2801.4, 2024: 3498.6},
-        "peak_year": 2024, "chronic_km2": 891.3, "chronic_pop": 1_200_000,
-        "chronic_districts": ["Kendrapara", "Jagatsinghpur", "Cuttack", "Puri"],
-        "high_risk_zones": ["Mahanadi delta coastal belt", "Kendrapara mangrove zone", "Chilika lake periphery", "Cuttack riverine islands"],
-        "accuracy_pct": 90.12, "risk_zones_km2": {"high": 2104.3, "moderate": 4312.7, "low": 6801.4},
-        "peak_rainfall_mm": 1350, "sar_threshold_db": -16,
     },
     "krishna": {
-        "name": "Krishna River Basin", "state": "Andhra Pradesh", "river": "Krishna",
-        "flood_areas": {2022: 2418.9, 2023: 1934.2, 2024: 2701.5},
-        "peak_year": 2024, "chronic_km2": 612.7, "chronic_pop": 820_000,
-        "chronic_districts": ["Krishna", "Guntur", "West Godavari", "Prakasam"],
-        "high_risk_zones": ["Krishna delta estuary", "Guntur low-lying agricultural belt", "Nagarjuna Sagar reservoir downstream", "Krishna-Godavari confluence zone"],
-        "accuracy_pct": 88.91, "risk_zones_km2": {"high": 1502.4, "moderate": 3214.8, "low": 5401.3},
-        "peak_rainfall_mm": 980, "sar_threshold_db": -16,
     },
     "godavari": {
-        "name": "Godavari Basin", "state": "Telangana / Andhra Pradesh", "river": "Godavari",
-        "flood_areas": {2022: 3814.2, 2023: 2612.8, 2024: 3109.4},
-        "peak_year": 2022, "chronic_km2": 1051.8, "chronic_pop": 1_500_000,
-        "chronic_districts": ["Bhadradri Kothagudem", "Mulugu", "East Godavari", "West Godavari"],
-        "high_risk_zones": ["Godavari riverine forest belt", "Bhadrachalam flood plains", "Papikonda gorge downstream", "East Godavari delta"],
-        "accuracy_pct": 91.08, "risk_zones_km2": {"high": 2401.7, "moderate": 4812.3, "low": 7204.6},
-        "peak_rainfall_mm": 1100, "sar_threshold_db": -16,
     },
 }
 DEFAULT_REGION = "brahmaputra"
-class BaseTask(ABC):
-    task_id:        str
-    name:           str
-    difficulty:     str
-    max_steps:      int
-    available_data: List[str]
-    def __init__(self, gee_available: bool = False, region: str = DEFAULT_REGION):
         self.gee_available = gee_available
         self.region_id     = region if region in REGIONS else DEFAULT_REGION
         self.region        = REGIONS[self.region_id]
-    @property
-    def description(self) -> str:
-        return self._make_description()
-    @abstractmethod
-    def _make_description(self) -> str: ...
-    @abstractmethod
-    def step(self, action: str, step_num: int) -> Dict[str, Any]: ...
     def get_context(self) -> Dict[str, Any]:
         r = self.region
         return {
-            "region": r["name"], "state": r["state"], "river": r["river"],
-            "years_available": [2022, 2023, 2024],
             "sar_threshold_db": r["sar_threshold_db"],
-            "flood_areas_km2": {str(k): v for k, v in r["flood_areas"].items()},
-            "peak_year": r["peak_year"],
         }
 class FloodYearComparisonTask(BaseTask):
-    task_id    = "flood_year_comparison"
-    name       = "Flood Year Comparison"
-    difficulty = "easy"
-    max_steps  = 6
-    available_data = ["Sentinel-1 SAR VV (2022-2024 June-Sept)", "CHIRPS rainfall (2022-2024)", "HydroSHEDS flow accumulation", "SRTM DEM"]
-    def __init__(self, **kwargs):
-        super().__init__(**kwargs)
-        self._rewarded_year   = False
-        self._rewarded_areas  = False
-        self._rewarded_reason = False
-    def _make_description(self) -> str:
         r = self.region
-        return (f"Using Sentinel-1 SAR data for the {r['name']} ({r['state']}), "
-                f"determine which monsoon year (2022-2024) had the LARGEST flood extent "
-                f"and report the area in square kilometres for all three years. "
-                f"Explain what drove the difference.")
-    def step(self, action: str, step_num: int) -> Dict[str, Any]:
-        txt    = action.lower()
-        r      = self.region
-        reward = 0.0
-        notes  = []
-        if not self._rewarded_year:
-            peak = str(r["peak_year"])
-            if peak in txt and any(w in txt for w in ["highest","largest","most","greatest","worst","maximum","peak","biggest","severe"]):
-                reward += 0.40
-                self._rewarded_year = True
-                notes.append(f"Correct peak year {peak} (+0.40)")
-        if not self._rewarded_areas:
-            years_hit = sum(1 for yr in [2022, 2023, 2024] if str(yr) in action)
-            nums      = _extract_nums(action)
-            close     = sum(1 for yr_a in r["flood_areas"].values()
-                           for n in nums if abs(n - yr_a) / yr_a < 0.15)
-            if years_hit >= 3 and close >= 2:
-                reward += 0.35
-                self._rewarded_areas = True
-                notes.append("All 3 year areas reported (+0.35)")
-            elif years_hit >= 2 and close >= 1:
-                reward += 0.15
-                notes.append("Partial areas (+0.15)")
-        if not self._rewarded_reason:
-            causal = ["rainfall","chirps","precipitation","monsoon","flow","accumulation","dem","elevation","slope","drainage","basin"]
-            if sum(1 for kw in causal if kw in txt) >= 2:
-                reward += 0.25
-                self._rewarded_reason = True
-                notes.append("Causal explanation (+0.25)")
-        done = (self._rewarded_year and self._rewarded_areas and self._rewarded_reason) or step_num >= self.max_steps
-        return {"reward": float(max(0.01, min(reward, 0.99))), "done": done,
-                "result": " | ".join(notes) if notes else "No criteria met.", "error": None}
     def get_context(self) -> Dict[str, Any]:
         ctx = super().get_context()
-        ctx["hint"] = f"Compare flood extents for 2022, 2023, 2024 in the {self.region['name']}."
         return ctx
-class DistrictInundationTask(BaseTask):
-    task_id    = "district_inundation_report"
-    name       = "Chronic District Inundation Report"
-    difficulty = "medium"
-    max_steps  = 8
-    available_data = ["Sentinel-1 SAR flood extents 2022-2024", "District boundaries (FAO GAUL)", "Flood frequency raster (0-3 years)", "WorldPop population grid", "NDWI permanent water mask"]
-    def __init__(self, **kwargs):
-        super().__init__(**kwargs)
-        self._found: set    = set()
-        self._rewarded_area = False
-        self._rewarded_pop  = False
-    def _make_description(self) -> str:
-        r = self.region
-        return (f"Using flood frequency analysis (2022-2024) for the {r['name']} ({r['state']}), "
-                f"identify which districts have been CHRONICALLY INUNDATED (flooded all 3 years). "
-                f"Report the total chronic area (km2) and estimate the affected population.")
-    def step(self, action: str, step_num: int) -> Dict[str, Any]:
-        txt    = action.lower()
-        r      = self.region
-        reward = 0.0
-        notes  = []
-        for district in r["chronic_districts"]:
-            dk = district.lower()
-            if dk not in self._found and dk in txt:
-                self._found.add(dk)
-                reward += 0.10
-                notes.append(f"District: {district} (+0.10)")
-        if not self._rewarded_area:
-            nums = _extract_nums(action)
-            if any(abs(n - r["chronic_km2"]) / r["chronic_km2"] < 0.20 for n in nums):
-                reward += 0.25
-                self._rewarded_area = True
-                notes.append(f"Chronic area ~{r['chronic_km2']} km2 (+0.25)")
-        if not self._rewarded_pop:
-            big = [n for n in _extract_nums(action) if n >= 100000]
-            if any(abs(n - r["chronic_pop"]) / r["chronic_pop"] < 0.30 for n in big):
-                reward += 0.25
-                self._rewarded_pop = True
-                notes.append("Population estimate (+0.25)")
-        done = (len(self._found) == len(r["chronic_districts"]) and
-                self._rewarded_area and self._rewarded_pop) or step_num >= self.max_steps
-        return {"reward": float(max(0.01, min(reward, 0.99))), "done": done,
-                "result": " | ".join(notes) if notes else f"Districts found: {list(self._found)}", "error": None}
-    def get_context(self) -> Dict[str, Any]:
-        ctx = super().get_context()
-        ctx.update({"target_districts": self.region["chronic_districts"],
-                    "chronic_area_km2": self.region["chronic_km2"],
-                    "approx_population": self.region["chronic_pop"]})
-        return ctx
 class FloodRiskForecastTask(BaseTask):
-    task_id    = "flood_risk_forecast"
-    name       = "Next-Season Flood Risk Forecast"
-    difficulty = "hard"
-    max_steps  = 10
-    available_data = ["Sentinel-1 SAR 2022-2024 flood extents", "CHIRPS rainfall trends",
-                      "Flood frequency map (0-3 years)", "Multi-factor risk zones",
-                      "Accuracy assessment metrics", "SRTM DEM + slope + HydroSHEDS"]
-    def __init__(self, **kwargs):
-        super().__init__(**kwargs)
-        self._acc   = False
-        self._zones = False
-        self._named = 0
-        self._rain  = False
-        self._year  = False
-    def _make_description(self) -> str:
-        r = self.region
-        return (f"Based on SAR flood history (2022-2024), rainfall trends, and the multi-factor "
-                f"risk model ({r['accuracy_pct']}% accuracy) for the {r['name']} ({r['state']}), "
-                f"forecast which zones face HIGHEST flood risk in the 2025 monsoon season.")
-    def step(self, action: str, step_num: int) -> Dict[str, Any]:
-        txt    = action.lower()
-        r      = self.region
-        reward = 0.0
-        notes  = []
-        if not self._acc:
-            acc_str = str(round(r["accuracy_pct"], 1))
-            if acc_str in action or any(k in txt for k in ["precision","recall","f1","accuracy"]):
-                reward += 0.15; self._acc = True
-                notes.append("Accuracy cited (+0.15)")
-        if not self._zones:
-            hits = sum(1 for kw in ["high risk","high-risk","moderate risk","low risk"] if kw in txt)
-            nums = _extract_nums(action)
-            hits += sum(1 for v in r["risk_zones_km2"].values()
-                       for n in nums if abs(n - v) / v < 0.05)
-            if hits >= 2:
-                reward += 0.20; self._zones = True
-                notes.append("Risk zones cited (+0.20)")
-            elif hits == 1:
-                reward += 0.08
-                notes.append("Partial zones (+0.08)")
-        if self._named < 2:
-            for zone in r["high_risk_zones"]:
-                if zone.lower() in txt and self._named < 2:
-                    self._named += 1; reward += 0.10
-                    notes.append(f"Zone: {zone} (+0.10)")
-        if not self._rain:
-            if any(k in txt for k in ["rainfall","chirps","precipitation","mm",str(r["peak_rainfall_mm"])]):
-                reward += 0.15; self._rain = True
-                notes.append("Rainfall data (+0.15)")
-        if not self._year:
-            peak = str(r["peak_year"])
-            if peak in txt and any(w in txt for w in ["worst","baseline","reference","peak","benchmark"]):
-                reward += 0.10; self._year = True
-                notes.append(f"{peak} as benchmark (+0.10)")
-        if "2025" in txt and any(w in txt for w in ["forecast","predict","expect","risk","likely"]):
-            reward += 0.05
-            notes.append("2025 forecast (+0.05)")
-        if not notes and len(action) < 120:
-            reward -= 0.10
-            notes.append("Too vague (-0.10)")
-        reward = max(reward, 0.0)
-        criteria = self._acc + self._zones + (self._named >= 2) + self._rain + self._year
-        done = criteria >= 4 or step_num >= self.max_steps
-        return {"reward": float(max(0.01, min(reward, 0.99))), "done": done,
-                "result": " | ".join(notes) if notes else "No criteria met.", "error": None}
     def get_context(self) -> Dict[str, Any]:
         ctx = super().get_context()
         r = self.region
         ctx.update({
             "model_accuracy_pct": r["accuracy_pct"],
-            "risk_zones_km2": r["risk_zones_km2"],
-            "high_risk_zones": r["high_risk_zones"],
-            "peak_rainfall_mm": r["peak_rainfall_mm"],
         })
         return ctx
 TASK_REGISTRY: Dict[str, type] = {
     "flood_year_comparison":      FloodYearComparisonTask,
-    "district_inundation_report": DistrictInundationTask,
     "flood_risk_forecast":        FloodRiskForecastTask,
 }

 """
+tasks.py — Chronostasis OpenEnv Task Definitions
+=================================================
+Multi-region flood intelligence environment for Indian river basins.
+Expanded from 5 to 15 basins covering ~85% of India's flood-prone population.
+Regions covered:
+  Original 5:  Brahmaputra, Ganga, Mahanadi, Krishna, Godavari
+  New 10:      Indus, Narmada, Tapti, Cauvery, Damodar,
+               Sabarmati, Mahi, Baitarani, Subarnarekha, Luni
 """
 import re
 from typing import Any, Dict, List, Optional
+# ──────────────────────────────────────────────
+# REGION DATA — 15 Indian river basins
+# Each region has lat/lon for map display,
+# seasonal risk multipliers, and full flood data.
+# ──────────────────────────────────────────────
+REGIONS: Dict[str, Dict[str, Any]] = {
+    # ── ORIGINAL 5 ───────────────────────────────────────────
     "brahmaputra": {
+        "name": "Brahmaputra Valley",
+        "state": "Assam",
+        "river": "Brahmaputra",
+        "lat": 26.2, "lon": 91.7,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 4812.3, 2023: 3601.7, 2024: 4102.8},
+        "peak_year": 2022,
+        "chronic_km2": 1823.4,
+        "chronic_pop": 2300000,
+        "chronic_districts": ["Dhubri", "Morigaon", "Barpeta", "Goalpara", "Kamrup"],
+        "high_risk_zones": ["Lower Assam Plains", "Brahmaputra Floodplain"],
+        "accuracy_pct": 92.39,
+        "risk_zones_km2": {"high": 3218.4, "moderate": 5901.2, "low": 8234.7},
+        "peak_rainfall_mm": 1587,
+        "seasonal_risk": {
+            "pre_monsoon": 0.3, "kharif": 0.95, "post_monsoon": 0.6, "rabi": 0.1
+        },
     },
     "ganga": {
+        "name": "Ganga Plains",
+        "state": "Bihar / UP",
+        "river": "Ganga",
+        "lat": 25.6, "lon": 85.1,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 3821.4, 2023: 4501.2, 2024: 3102.6},
+        "peak_year": 2023,
+        "chronic_km2": 2103.6,
+        "chronic_pop": 3100000,
+        "chronic_districts": ["Patna", "Bhagalpur", "Darbhanga", "Muzaffarpur", "Samastipur"],
+        "high_risk_zones": ["North Bihar Plains", "Kosi Fan"],
+        "accuracy_pct": 89.7,
+        "risk_zones_km2": {"high": 2914.8, "moderate": 6203.1, "low": 9401.5},
+        "peak_rainfall_mm": 1423,
+        "seasonal_risk": {
+            "pre_monsoon": 0.2, "kharif": 0.90, "post_monsoon": 0.5, "rabi": 0.1
+        },
     },
     "mahanadi": {
+        "name": "Mahanadi Delta",
+        "state": "Odisha",
+        "river": "Mahanadi",
+        "lat": 20.5, "lon": 85.8,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 2914.7, 2023: 2103.8, 2024: 3401.5},
+        "peak_year": 2024,
+        "chronic_km2": 1402.3,
+        "chronic_pop": 1800000,
+        "chronic_districts": ["Cuttack", "Kendrapara", "Jagatsinghpur", "Puri", "Khordha"],
+        "high_risk_zones": ["Mahanadi Delta", "Coastal Odisha"],
+        "accuracy_pct": 90.1,
+        "risk_zones_km2": {"high": 2103.4, "moderate": 4801.2, "low": 7203.8},
+        "peak_rainfall_mm": 1312,
+        "seasonal_risk": {
+            "pre_monsoon": 0.25, "kharif": 0.88, "post_monsoon": 0.55, "rabi": 0.1
+        },
     },
     "krishna": {
+        "name": "Krishna River Basin",
+        "state": "Andhra Pradesh",
+        "river": "Krishna",
+        "lat": 16.5, "lon": 80.6,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 1823.5, 2023: 2914.2, 2024: 1502.8},
+        "peak_year": 2023,
+        "chronic_km2": 892.1,
+        "chronic_pop": 1200000,
+        "chronic_districts": ["Guntur", "Krishna", "West Godavari", "Prakasam", "Nalgonda"],
+        "high_risk_zones": ["Krishna Delta", "Lower Krishna Plains"],
+        "accuracy_pct": 88.9,
+        "risk_zones_km2": {"high": 1402.3, "moderate": 3201.8, "low": 5803.4},
+        "peak_rainfall_mm": 1089,
+        "seasonal_risk": {
+            "pre_monsoon": 0.2, "kharif": 0.85, "post_monsoon": 0.65, "rabi": 0.15
+        },
     },
     "godavari": {
+        "name": "Godavari Basin",
+        "state": "Telangana / AP",
+        "river": "Godavari",
+        "lat": 17.0, "lon": 81.8,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 3102.4, 2023: 2801.6, 2024: 3891.3},
+        "peak_year": 2024,
+        "chronic_km2": 1601.2,
+        "chronic_pop": 2100000,
+        "chronic_districts": ["East Godavari", "West Godavari", "Khammam", "Bhadradri", "Devanahalli"],
+        "high_risk_zones": ["Godavari Delta", "Lower Godavari Plains"],
+        "accuracy_pct": 91.1,
+        "risk_zones_km2": {"high": 2401.6, "moderate": 5102.3, "low": 7803.9},
+        "peak_rainfall_mm": 1198,
+        "seasonal_risk": {
+            "pre_monsoon": 0.25, "kharif": 0.87, "post_monsoon": 0.60, "rabi": 0.12
+        },
+    },
+    # ── NEW REGIONS ───────────────────────────────────────────
+    "narmada": {
+        "name": "Narmada Basin",
+        "state": "Madhya Pradesh / Gujarat",
+        "river": "Narmada",
+        "lat": 22.7, "lon": 77.4,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 1823.6, 2023: 2401.3, 2024: 1602.8},
+        "peak_year": 2023,
+        "chronic_km2": 892.4,
+        "chronic_pop": 1100000,
+        "chronic_districts": ["Hoshangabad", "Jabalpur", "Narsinghpur", "Bharuch", "Narmadapuram"],
+        "high_risk_zones": ["Narmada Valley", "Sardar Sarovar Backwaters"],
+        "accuracy_pct": 87.3,
+        "risk_zones_km2": {"high": 1203.4, "moderate": 2801.2, "low": 4903.6},
+        "peak_rainfall_mm": 1134,
+        "seasonal_risk": {
+            "pre_monsoon": 0.15, "kharif": 0.82, "post_monsoon": 0.45, "rabi": 0.08
+        },
+    },
+    "tapti": {
+        "name": "Tapti Basin",
+        "state": "Maharashtra / Gujarat",
+        "river": "Tapti",
+        "lat": 21.2, "lon": 74.8,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 1203.4, 2023: 1801.2, 2024: 1402.6},
+        "peak_year": 2023,
+        "chronic_km2": 601.3,
+        "chronic_pop": 780000,
+        "chronic_districts": ["Surat", "Tapi", "Nandurbar", "Dhule", "Jalgaon"],
+        "high_risk_zones": ["Surat Lowlands", "Tapti Floodplain"],
+        "accuracy_pct": 86.8,
+        "risk_zones_km2": {"high": 801.4, "moderate": 1802.3, "low": 3201.5},
+        "peak_rainfall_mm": 987,
+        "seasonal_risk": {
+            "pre_monsoon": 0.12, "kharif": 0.80, "post_monsoon": 0.40, "rabi": 0.07
+        },
+    },
+    "cauvery": {
+        "name": "Cauvery Basin",
+        "state": "Karnataka / Tamil Nadu",
+        "river": "Cauvery",
+        "lat": 12.3, "lon": 77.0,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 1102.3, 2023: 1503.8, 2024: 1301.4},
+        "peak_year": 2023,
+        "chronic_km2": 542.1,
+        "chronic_pop": 890000,
+        "chronic_districts": ["Thanjavur", "Tiruvarur", "Nagapattinam", "Mysuru", "Mandya"],
+        "high_risk_zones": ["Cauvery Delta", "Thanjavur Plains"],
+        "accuracy_pct": 88.2,
+        "risk_zones_km2": {"high": 703.4, "moderate": 1601.2, "low": 2903.8},
+        "peak_rainfall_mm": 892,
+        "seasonal_risk": {
+            "pre_monsoon": 0.18, "kharif": 0.75, "post_monsoon": 0.70, "rabi": 0.20
+        },
+    },
+    "damodar": {
+        "name": "Damodar Valley",
+        "state": "Jharkhand / West Bengal",
+        "river": "Damodar",
+        "lat": 23.5, "lon": 87.3,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 2103.4, 2023: 1801.6, 2024: 2401.8},
+        "peak_year": 2024,
+        "chronic_km2": 1012.3,
+        "chronic_pop": 1400000,
+        "chronic_districts": ["Barddhaman", "Hooghly", "Howrah", "Dhanbad", "Bokaro"],
+        "high_risk_zones": ["Damodar Floodplain", "Lower Damodar Valley"],
+        "accuracy_pct": 89.4,
+        "risk_zones_km2": {"high": 1401.3, "moderate": 3201.8, "low": 5102.4},
+        "peak_rainfall_mm": 1203,
+        "seasonal_risk": {
+            "pre_monsoon": 0.20, "kharif": 0.88, "post_monsoon": 0.50, "rabi": 0.10
+        },
+    },
+    "sabarmati": {
+        "name": "Sabarmati Basin",
+        "state": "Gujarat / Rajasthan",
+        "river": "Sabarmati",
+        "lat": 23.0, "lon": 72.6,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 801.3, 2023: 1203.4, 2024: 902.6},
+        "peak_year": 2023,
+        "chronic_km2": 312.4,
+        "chronic_pop": 420000,
+        "chronic_districts": ["Ahmedabad", "Gandhinagar", "Mehsana", "Sabarkantha", "Patan"],
+        "high_risk_zones": ["Ahmedabad Lowlands", "Sabarmati Floodplain"],
+        "accuracy_pct": 85.6,
+        "risk_zones_km2": {"high": 401.2, "moderate": 901.4, "low": 1803.8},
+        "peak_rainfall_mm": 734,
+        "seasonal_risk": {
+            "pre_monsoon": 0.10, "kharif": 0.75, "post_monsoon": 0.30, "rabi": 0.05
+        },
+    },
+    "mahi": {
+        "name": "Mahi Basin",
+        "state": "Gujarat / Rajasthan / MP",
+        "river": "Mahi",
+        "lat": 22.8, "lon": 73.5,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 712.3, 2023: 1103.4, 2024: 834.6},
+        "peak_year": 2023,
+        "chronic_km2": 298.7,
+        "chronic_pop": 380000,
+        "chronic_districts": ["Vadodara", "Anand", "Kheda", "Panchmahal", "Dahod"],
+        "high_risk_zones": ["Mahi Delta", "Vadodara Lowlands"],
+        "accuracy_pct": 84.9,
+        "risk_zones_km2": {"high": 312.4, "moderate": 801.2, "low": 1602.8},
+        "peak_rainfall_mm": 812,
+        "seasonal_risk": {
+            "pre_monsoon": 0.10, "kharif": 0.78, "post_monsoon": 0.35, "rabi": 0.06
+        },
+    },
+    "baitarani": {
+        "name": "Baitarani Basin",
+        "state": "Odisha / Jharkhand",
+        "river": "Baitarani",
+        "lat": 21.5, "lon": 86.4,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 1203.4, 2023: 1601.8, 2024: 1401.2},
+        "peak_year": 2023,
+        "chronic_km2": 612.3,
+        "chronic_pop": 820000,
+        "chronic_districts": ["Bhadrak", "Jajpur", "Kendujhar", "Balasore", "Mayurbhanj"],
+        "high_risk_zones": ["Baitarani Delta", "Lower Odisha Coast"],
+        "accuracy_pct": 87.1,
+        "risk_zones_km2": {"high": 801.4, "moderate": 1802.3, "low": 3201.5},
+        "peak_rainfall_mm": 1089,
+        "seasonal_risk": {
+            "pre_monsoon": 0.22, "kharif": 0.85, "post_monsoon": 0.55, "rabi": 0.10
+        },
+    },
+    "subarnarekha": {
+        "name": "Subarnarekha Basin",
+        "state": "Jharkhand / WB / Odisha",
+        "river": "Subarnarekha",
+        "lat": 22.3, "lon": 86.9,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 912.3, 2023: 1203.4, 2024: 1034.6},
+        "peak_year": 2023,
+        "chronic_km2": 412.8,
+        "chronic_pop": 560000,
+        "chronic_districts": ["East Singhbhum", "West Midnapore", "Balasore", "Seraikela", "Kharsawan"],
+        "high_risk_zones": ["Subarnarekha Delta", "Jamshedpur Lowlands"],
+        "accuracy_pct": 86.3,
+        "risk_zones_km2": {"high": 601.4, "moderate": 1301.2, "low": 2401.6},
+        "peak_rainfall_mm": 1134,
+        "seasonal_risk": {
+            "pre_monsoon": 0.20, "kharif": 0.83, "post_monsoon": 0.50, "rabi": 0.09
+        },
+    },
+    "indus": {
+        "name": "Indus Plains",
+        "state": "Punjab / Haryana / J&K",
+        "river": "Indus / Sutlej",
+        "lat": 30.9, "lon": 75.8,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 2301.4, 2023: 1803.6, 2024: 2103.8},
+        "peak_year": 2022,
+        "chronic_km2": 1102.3,
+        "chronic_pop": 1600000,
+        "chronic_districts": ["Ludhiana", "Jalandhar", "Amritsar", "Firozpur", "Fazilka"],
+        "high_risk_zones": ["Punjab Doab", "Sutlej Floodplain"],
+        "accuracy_pct": 88.4,
+        "risk_zones_km2": {"high": 1503.4, "moderate": 3401.2, "low": 5803.6},
+        "peak_rainfall_mm": 812,
+        "seasonal_risk": {
+            "pre_monsoon": 0.15, "kharif": 0.80, "post_monsoon": 0.40, "rabi": 0.08
+        },
+    },
+    "luni": {
+        "name": "Luni Basin",
+        "state": "Rajasthan / Gujarat",
+        "river": "Luni",
+        "lat": 25.8, "lon": 72.1,
+        "sar_threshold_db": -16,
+        "flood_areas": {2022: 612.3, 2023: 1203.4, 2024: 803.6},
+        "peak_year": 2023,
+        "chronic_km2": 231.4,
+        "chronic_pop": 290000,
+        "chronic_districts": ["Barmer", "Jalor", "Pali", "Jodhpur", "Sirohi"],
+        "high_risk_zones": ["Luni Floodplain", "Barmer Lowlands"],
+        "accuracy_pct": 83.7,
+        "risk_zones_km2": {"high": 301.2, "moderate": 703.4, "low": 1402.8},
+        "peak_rainfall_mm": 412,
+        "seasonal_risk": {
+            "pre_monsoon": 0.05, "kharif": 0.70, "post_monsoon": 0.20, "rabi": 0.03
+        },
     },
 }
 DEFAULT_REGION = "brahmaputra"
+# Seasonal descriptions for context
+SEASON_DESCRIPTIONS = {
+    "pre_monsoon":  "March–May: dry season, low base flow, localised storm risk",
+    "kharif":       "June–September: peak monsoon, maximum flood risk",
+    "post_monsoon": "October–November: receding waters, secondary flood risk",
+    "rabi":         "December–February: winter season, minimal flood risk",
+}
+# ──────────────────────────────────────────────
+# BASE TASK
+# ──────────────────────────────────────────────
+class BaseTask:
+    task_id:      str = ""
+    name:         str = ""
+    description:  str = ""
+    difficulty:   str = "easy"
+    max_steps:    int = 6
+    available_data: List[str] = []
+    def __init__(self, gee_available: bool = False,
+                 region: str = DEFAULT_REGION,
+                 season: str = "kharif"):
         self.gee_available = gee_available
         self.region_id     = region if region in REGIONS else DEFAULT_REGION
         self.region        = REGIONS[self.region_id]
+        self.season        = season if season in SEASON_DESCRIPTIONS else "kharif"
     def get_context(self) -> Dict[str, Any]:
         r = self.region
+        fa = r["flood_areas"]
         return {
+            "region":           r["name"],
+            "state":            r["state"],
+            "river":            r["river"],
+            "lat":              r["lat"],
+            "lon":              r["lon"],
+            "years_available":  sorted(fa.keys()),
+            "flood_areas_km2":  fa,
             "sar_threshold_db": r["sar_threshold_db"],
+            "peak_year":        r["peak_year"],
+            "season":           self.season,
+            "season_desc":      SEASON_DESCRIPTIONS[self.season],
+            "seasonal_risk":    r["seasonal_risk"][self.season],
+            "hint":             f"Compare flood extents for {', '.join(str(y) for y in sorted(fa.keys()))} in the {r['name']}.",
         }
+    def step(self, response: str, step_num: int) -> Dict[str, Any]:
+        raise NotImplementedError
+# ──────────────────────────────────────────────
+# REWARD HELPERS
+# ──────────────────────────────────────────────
+def _clamp(v: float) -> float:
+    """Reward must be strictly between 0 and 1."""
+    return max(0.01, min(float(v), 0.99))
+def _extract_numbers(text: str) -> List[float]:
+    return [float(x.replace(",", "")) for x in re.findall(r"\d[\d,]*\.?\d*", text)]
+def _mentions_any(text: str, terms: List[str]) -> bool:
+    tl = text.lower()
+    return any(t.lower() in tl for t in terms)
+def _penalty_vague(text: str) -> float:
+    vague_phrases = [
+        "some areas", "many districts", "various regions",
+        "flood prone", "several years", "significant flooding",
+        "major impact", "affected areas", "heavy rainfall",
+        "flood risk exists",
+    ]
+    hits = sum(1 for p in vague_phrases if p in text.lower())
+    return -0.10 * min(hits, 3)
+def _causal_score(text: str) -> float:
+    causal_terms = ["chirps", "dem", "slope", "hydrosheds", "flow accumulation",
+                    "sar", "sentinel", "elevation", "drainage", "catchment",
+                    "rainfall", "discharge", "ndwi", "worldpop"]
+    hits = sum(1 for t in causal_terms if t in text.lower())
+    return min(hits * 0.05, 0.20)
+# ──────────────────────────────────────────────
+# TASK 1 — EASY
+# ──────────────────────────────────────────────
 class FloodYearComparisonTask(BaseTask):
+    task_id     = "flood_year_comparison"
+    name        = "SAR Flood Year Comparison"
+    description = (
+        "Using Sentinel-1 SAR data, determine which monsoon year (2022–2024) "
+        "had the LARGEST flood extent and report the area in square kilometres "
+        "for all three years. Explain what drove the difference."
+    )
+    difficulty  = "easy"
+    max_steps   = 6
+    available_data = [
+        "Sentinel-1 SAR VV (2022–2024 June–Sept)",
+        "CHIRPS daily rainfall (2022–2024)",
+        "HydroSHEDS flow accumulation (15ACC)",
+        "SRTM DEM (30m resolution)",
+        "Landsat 8 NDWI permanent water mask",
+    ]
+    def get_context(self) -> Dict[str, Any]:
+        ctx = super().get_context()
         r = self.region
+        fa = r["flood_areas"]
+        ctx.update({
+            "flood_areas_km2": fa,
+            "peak_year":       r["peak_year"],
+        })
+        return ctx
+    def step(self, response: str, step_num: int) -> Dict[str, Any]:
+        r    = self.region
+        fa   = r["flood_areas"]
+        nums = _extract_numbers(response)
+        score = 0.0
+        # Year identification
+        peak = r["peak_year"]
+        if str(peak) in response:
+            score += 0.30
+        # Numeric accuracy — check all 3 years
+        for yr, area in fa.items():
+            for n in nums:
+                if abs(n - area) / area < 0.05:
+                    score += 0.15
+                    break
+        # Causal explanation
+        score += _causal_score(response)
+        # Vague penalty
+        score += _penalty_vague(response)
+        done = step_num >= self.max_steps
+        return {"reward": _clamp(score), "done": done,
+                "result": f"Step {step_num}: scored {score:.3f}"}
+# ──────────────────────────────────────────────
+# TASK 2 — MEDIUM
+# ──────────────────────────────────────────────
+class DistrictInundationReportTask(BaseTask):
+    task_id     = "district_inundation_report"
+    name        = "District Chronic Inundation Report"
+    description = (
+        "Identify districts with CHRONIC inundation (flooded in all 3 years: "
+        "2022, 2023, 2024). Report the total chronically inundated area in km², "
+        "the estimated affected population, and the primary causal factors "
+        "for each district's recurring flood vulnerability."
+    )
+    difficulty  = "medium"
+    max_steps   = 8
+    available_data = [
+        "Sentinel-1 SAR VV (2022–2024 June–Sept)",
+        "CHIRPS daily rainfall (2022–2024)",
+        "HydroSHEDS flow accumulation (15ACC)",
+        "SRTM DEM (30m resolution)",
+        "FAO GAUL district boundaries",
+        "WorldPop population density (2020)",
+        "Landsat 8 NDWI permanent water mask",
+    ]
     def get_context(self) -> Dict[str, Any]:
         ctx = super().get_context()
+        r = self.region
+        ctx.update({
+            "chronic_area_km2":  r["chronic_km2"],
+            "chronic_population": r["chronic_pop"],
+            "target_districts":  r["chronic_districts"],
+        })
         return ctx
+    def step(self, response: str, step_num: int) -> Dict[str, Any]:
+        r    = self.region
+        nums = _extract_numbers(response)
+        score = 0.0
+        # District names
+        hit_districts = sum(1 for d in r["chronic_districts"] if d.lower() in response.lower())
+        score += min(hit_districts * 0.12, 0.36)
+        # Chronic area
+        for n in nums:
+            if abs(n - r["chronic_km2"]) / r["chronic_km2"] < 0.10:
+                score += 0.20
+                break
+        # Population
+        pop_millions = r["chronic_pop"] / 1e6
+        for n in nums:
+            if abs(n - r["chronic_pop"]) / r["chronic_pop"] < 0.15 or \
+               abs(n - pop_millions) / pop_millions < 0.15:
+                score += 0.15
+                break
+        # Causal
+        score += _causal_score(response)
+        # Vague penalty
+        score += _penalty_vague(response)
+        done = step_num >= self.max_steps
+        return {"reward": _clamp(score), "done": done,
+                "result": f"Step {step_num}: scored {score:.3f}"}
+# ──────────────────────────────────────────────
+# TASK 3 — HARD
+# ──────────────────────────────────────────────
 class FloodRiskForecastTask(BaseTask):
+    task_id     = "flood_risk_forecast"
+    name        = "2025 Monsoon Flood Risk Forecast"
+    description = (
+        "Using the multi-factor risk model (92%+ accuracy), forecast the "
+        "HIGH-RISK flood zones for the 2025 monsoon season. Report zone areas "
+        "in km², identify specific geographic zones by name, cite the causal "
+        "factors (CHIRPS trend, DEM, slope, flow accumulation), and recommend "
+        "early warning priorities."
+    )
+    difficulty  = "hard"
+    max_steps   = 10
+    available_data = [
+        "Sentinel-1 SAR VV (2022–2024 June–Sept)",
+        "CHIRPS daily rainfall + 10-year trend (2015–2024)",
+        "HydroSHEDS flow accumulation (15ACC)",
+        "SRTM DEM + slope (30m resolution)",
+        "FAO GAUL district boundaries",
+        "WorldPop population density (2020)",
+        "Landsat 8 NDWI permanent water mask",
+        "Multi-factor risk model (SVM + Random Forest ensemble)",
+    ]
     def get_context(self) -> Dict[str, Any]:
         ctx = super().get_context()
         r = self.region
         ctx.update({
             "model_accuracy_pct": r["accuracy_pct"],
+            "risk_zones_km2":     r["risk_zones_km2"],
+            "high_risk_zones":    r["high_risk_zones"],
+            "peak_rainfall_mm":   r["peak_rainfall_mm"],
         })
         return ctx
+    def step(self, response: str, step_num: int) -> Dict[str, Any]:
+        r    = self.region
+        rz   = r["risk_zones_km2"]
+        nums = _extract_numbers(response)
+        score = 0.0
+        # Model accuracy cited
+        for n in nums:
+            if abs(n - r["accuracy_pct"]) < 2.0:
+                score += 0.15
+                break
+        # Risk zone areas
+        for zone_val in rz.values():
+            for n in nums:
+                if abs(n - zone_val) / zone_val < 0.08:
+                    score += 0.12
+                    break
+        # High-risk zone names
+        hit_zones = sum(1 for z in r["high_risk_zones"] if z.lower() in response.lower())
+        score += min(hit_zones * 0.10, 0.20)
+        # Causal factors
+        score += _causal_score(response)
+        # Early warning / recommendation language
+        if _mentions_any(response, ["early warning", "evacuate", "alert", "priority", "recommend"]):
+            score += 0.05
+        # Vague penalty
+        score += _penalty_vague(response)
+        done = step_num >= self.max_steps
+        return {"reward": _clamp(score), "done": done,
+                "result": f"Step {step_num}: scored {score:.3f}"}
+# ──────────────────────────────────────────────
+# REGISTRY
+# ──────────────────────────────────────────────
 TASK_REGISTRY: Dict[str, type] = {
     "flood_year_comparison":      FloodYearComparisonTask,
+    "district_inundation_report": DistrictInundationReportTask,
     "flood_risk_forecast":        FloodRiskForecastTask,
 }