surtgis 0.7.2

High-performance geospatial analysis library for Python, powered by Rust

surtgis

High-performance geospatial analysis library for Python, powered by Rust.

SurtGIS provides 40+ terrain, hydrology, and imagery algorithms with native performance through PyO3 bindings.

Installation

pip install surtgis

Quick Start

import numpy as np
import surtgis

# Create a sample DEM (100x100 grid)
dem = np.random.rand(100, 100) * 100  # Elevation in meters
cell_size = 30.0  # 30m resolution

# Compute terrain derivatives
slope = surtgis.slope(dem, cell_size)
aspect = surtgis.aspect_degrees(dem, cell_size)
hillshade = surtgis.hillshade_compute(dem, cell_size, azimuth=315.0, altitude=45.0)

# Curvature analysis
profile_curv = surtgis.curvature_compute(dem, cell_size, ctype="profile")
plan_curv = surtgis.curvature_compute(dem, cell_size, ctype="plan")

# Advanced Florinsky curvatures (14 types available)
mean_h = surtgis.advanced_curvature(dem, cell_size, ctype="mean_h")
gaussian_k = surtgis.advanced_curvature(dem, cell_size, ctype="gaussian_k")

# Hydrology
twi = surtgis.twi_compute(dem, cell_size)  # Topographic Wetness Index
hand = surtgis.hand_compute(dem, cell_size)  # Height Above Nearest Drainage

print(f"Slope range: {slope.min():.1f} - {slope.max():.1f}")
print(f"TWI range: {twi.min():.2f} - {twi.max():.2f}")

Available Functions

Terrain Analysis (21 functions)

Function	Description
slope(dem, cell_size)	Slope in degrees or percent
aspect_degrees(dem, cell_size)	Aspect in degrees (0-360)
hillshade_compute(dem, cell_size, azimuth, altitude)	Analytical hillshade
multidirectional_hillshade(dem, cell_size)	Multi-azimuth hillshade
curvature_compute(dem, cell_size, ctype)	Profile, plan, or general curvature
advanced_curvature(dem, cell_size, ctype)	14 Florinsky curvatures
tpi_compute(dem, cell_size, radius)	Topographic Position Index
tri_compute(dem, cell_size)	Terrain Ruggedness Index
dev_compute(dem, cell_size, radius)	Deviation from Mean Elevation
vrm_compute(dem, cell_size, radius)	Vector Ruggedness Measure
geomorphons_compute(dem, cell_size, flatness, radius)	Landform classification
mrvbf_compute(dem, cell_size)	Multi-resolution Valley Bottom Flatness
northness_compute(dem, cell_size)	Cosine of aspect
eastness_compute(dem, cell_size)	Sine of aspect
shape_index_compute(dem, cell_size)	Shape index
curvedness_compute(dem, cell_size)	Curvedness
sky_view_factor_compute(dem, cell_size)	Sky View Factor
viewshed_compute(dem, cell_size, observer_row, observer_col)	Line-of-sight visibility
openness_positive(dem, cell_size)	Positive openness
openness_negative(dem, cell_size)	Negative openness
uncertainty_slope(dem, cell_size, dem_rmse)	Slope uncertainty from DEM error

Hydrology (9 functions)

Function	Description
fill_depressions(dem, cell_size)	Planchon-Darboux sink filling
priority_flood_fill(dem, cell_size)	Priority-Flood (Barnes 2014)
breach_fill(dem, cell_size)	Breach depressions (Lindsay 2016)
flow_direction_d8(dem, cell_size)	D8 flow direction
flow_accumulation_d8(fdir, cell_size)	D8 flow accumulation
flow_accumulation_mfd_compute(dem, cell_size)	Multi-flow direction accumulation
twi_compute(dem, cell_size)	Topographic Wetness Index
hand_compute(dem, cell_size)	Height Above Nearest Drainage
stream_network_compute(dem, cell_size, threshold)	Stream network extraction

Imagery (4 functions)

Function	Description
ndvi_compute(nir, red)	Normalized Difference Vegetation Index
ndwi_compute(green, nir)	Normalized Difference Water Index
savi_compute(nir, red, l_factor)	Soil-Adjusted Vegetation Index
normalized_diff(a, b)	Generic normalized difference

Morphology (4 functions)

Function	Description
morph_erode(data, radius)	Morphological erosion
morph_dilate(data, radius)	Morphological dilation
morph_opening(data, radius)	Morphological opening
morph_closing(data, radius)	Morphological closing

Statistics (3 functions)

Function	Description
focal_mean(data, radius)	Focal mean
focal_std(data, radius)	Focal standard deviation
focal_range(data, radius)	Focal range (max - min)

Advanced Curvature Types

The advanced_curvature function supports 14 curvature types from Florinsky (2025):

• mean_h - Mean curvature H
• gaussian_k - Gaussian curvature K
• unsphericity_m - Unsphericity M
• difference_e - Difference curvature E
• minimal_kmin - Minimal principal curvature
• maximal_kmax - Maximal principal curvature
• horizontal_kh - Horizontal (plan) curvature
• vertical_kv - Vertical (profile) curvature
• horizontal_excess_khe - Horizontal excess curvature
• vertical_excess_kve - Vertical excess curvature
• accumulation_ka - Accumulation curvature
• ring_kr - Ring curvature
• rotor - Rotor (flow-line torsion)
• laplacian - Laplacian

Performance

SurtGIS is 5-10x faster than pure Python implementations thanks to Rust's zero-cost abstractions and automatic parallelization via Rayon.

License

MIT OR Apache-2.0