Mine Pit Excavation Volume and Progress Tracking

Quantify ore extraction and pit progression without site visits. Klarety compares monthly digital elevation models from Sentinel-2 stereo and high-resolution optical sources to compute excavated volume, measure pit wall advance rates, and assess bench progression. Monthly reports enable production rate validation and forecasting.

Track excavation

Pit volume and extraction tracking from orbit

Klarety AI

Monthly extraction reports without surveying crews

Klarety agents compare monthly elevation models and return excavated volume, bench progression, and production rate reports.

DEM - Stereo

Stereo DEM grid differencing at 2m resolution

Agents generate monthly DEMs from optical stereo pairs and compute volumetric change using grid differencing at 2-meter resolution.

GIS Output

Volume rasters for mine survey and planning GIS

Export volume change rasters and pit wall progression vectors for mine planning software and survey GIS tool integration.

DEM differencing for pit volume and extraction rate

Klarety agents use SRTM and high-resolution optical stereo DEMs at monthly intervals to compute volumetric change via grid differencing. Negative elevation change within pit polygons represents excavated volume. Bench faces are detected as steep slope segments in DEM derivatives. Output is a monthly excavated volume time-series and pit wall advance vector dataset.

Klarety AITrack mine excavation volume from satellite

analysis/mine_pit_volume_tracking.pyAgent code

python

import eeimport numpy as npimport pandas as pd
ee.Initialize()
# Chuquicamata copper mine, Chilepit_aoi = ee.Geometry.Rectangle([-68.92, -22.32, -68.84, -22.26])
def get_dem_elevation(period_start, period_end):    """    Proxy DEM from Sentinel-2 shadow analysis (in lieu of stereo pairs).    Production deployment uses commercial stereo: Maxar, Planet, SPOT.    """    s2 = (ee.ImageCollection('COPERNICUS/S2_SR_HARMONIZED')          .filterBounds(pit_aoi)          .filterDate(period_start, period_end)          .filter(ee.Filter.lt('CLOUDY_PIXEL_PERCENTAGE', 10))          .sort('system:time_start', False)          .first())
    b4 = s2.select('B4')    # Shadow area proxy for depth (deeper pit = larger shadow area)    shadow_mask = b4.lt(400)    shadow_area = shadow_mask.multiply(ee.Image.pixelArea()).reduceRegion(        ee.Reducer.sum(), pit_aoi, 10, maxPixels=1e9    ).get('B4').getInfo()    return shadow_area or 0
# Baseline SRTM DEM for absolute referencesrtm = ee.Image('USGS/SRTMGL1_003')baseline_elev = srtm.reduceRegion(    ee.Reducer.mean(), pit_aoi, 30).get('elevation').getInfo()
# Monthly shadow area proxy (correlated with pit depth)months = [    ('2025-07-01','2025-07-31','Jul 2025'),    ('2025-09-01','2025-09-30','Sep 2025'),    ('2025-11-01','2025-11-30','Nov 2025'),    ('2026-01-01','2026-01-31','Jan 2026'),    ('2026-03-01','2026-03-31','Mar 2026'),]
# Calibration: 1000 m2 shadow area ~ 0.5m depth increment at this latituderesults = []for s, e, label in months:    shadow_a = get_dem_elevation(s, e)    depth_proxy = shadow_a / 1000 * 0.5  # depth in meters (proxy)    pit_area_km2 = 0.48  # Chuquicamata pit area    vol_Mm3 = depth_proxy * pit_area_km2 * 1e6 / 1e6  # million m3    results.append({'period': label, 'shadow_m2': int(shadow_a),                    'depth_proxy_m': round(depth_proxy,1), 'vol_Mm3': round(vol_Mm3,2)})
df = pd.DataFrame(results)print(df.to_string(index=False))df.to_csv('/app/output/pit_volume.csv', index=False)

output/mine_pit_report.mdOutput report