Complete Guide to GIS Spectral Indices: NDVI, NDWI, NDBI & More

Index	Formula (generic bands)	Typical usage
NDVI	(NIR − RED) / (NIR + RED)	Vegetation vigor/greenness
EVI	2.5 × (NIR − RED) / (NIR + 6×RED − 7.5×BLUE + 1)	Vegetation (reduces soil/atmospheric effects)
SAVI	((NIR − RED) / (NIR + RED + L)) × (1 + L), L ≈ 0.5	Vegetation in sparse cover/soil backgrounds
MSAVI	(2×NIR + 1 − √((2×NIR + 1)² − 8×(NIR − RED))) / 2	Vegetation with minimal soil influence
GNDVI	(NIR − GREEN) / (NIR + GREEN)	Chlorophyll/vegetation stress
NDRE	(NIR − RE) / (NIR + RE)	Canopy chlorophyll (uses Red-Edge)
ARVI	(NIR − (2×RED − BLUE)) / (NIR + (2×RED − BLUE))	Vegetation with atmospheric correction
VARI	(GREEN − RED) / (GREEN + RED − BLUE)	Vegetation in RGB (low-cost imagery)
NDMI (a.k.a. NDWI-Gao)	(NIR − SWIR1) / (NIR + SWIR1)	Canopy moisture/forest drought
NDWI (McFeeters)	(GREEN − NIR) / (GREEN + NIR)	Open water extraction (may confuse built-up)
MNDWI (Xu)	(GREEN − SWIR1) / (GREEN + SWIR1)	Water extraction (better in urban)
NDBI	(SWIR1 − NIR) / (SWIR1 + NIR)	Built-up detection
BSI	((SWIR1 + RED) − (NIR + BLUE)) / ((SWIR1 + RED) + (NIR + BLUE))	Bare soil identification
NDSI (Snow)	(GREEN − SWIR1) / (GREEN + SWIR1)	Snow mapping
NBR	(NIR − SWIR2) / (NIR + SWIR2)	Burn severity/disturbance
NBR2	(SWIR1 − SWIR2) / (SWIR1 + SWIR2)	Moisture/thermal disturbance in dry regions
EBBI	(SWIR1 − NIR) / √(SWIR1 + TIR)	Built-up (requires thermal)
SIPI	(NIR − BLUE) / (NIR − RED)	Pigment stress (carotenoid vs. chlorophyll)
PRI	(GREEN531 − GREEN570) / (GREEN531 + GREEN570)	Photosynthetic efficiency (narrow bands)
GEMI	η(1 − 0.25η) − (RED − 0.125)/(1 − RED), where η = 2((NIR² − RED²) + 1.5NIR + 0.5RED)/(NIR + RED + 0.5)	Atmospheric-resilient greenness

Band keys (generic): BLUE (≈0.45–0.50 μm), GREEN (≈0.53–0.59 μm), RED (≈0.64–0.68 μm), RE = Red-Edge (≈0.70–0.74 μm), NIR (≈0.84–0.88 μm), SWIR₁ (≈1.57–1.65 μm), SWIR₂ (≈2.11–2.29 μm), TIR = Thermal (10–12 μm). For SAVI, a common soil factor is L = 0.5.

Band Numbers

Generic Name	Sentinel-2 Band	Landsat-8/9 Band	Wavelength (µm)
BLUE	B2	B2	0.45 – 0.50
GREEN	B3	B3	0.53 – 0.59
RED	B4	B4	0.64 – 0.68
Red-Edge (RE)	B5, B6, B7	–	0.70 – 0.74
NIR	B8	B5	0.84 – 0.88
SWIR1	B11	B6	1.57 – 1.65
SWIR2	B12	B7	2.11 – 2.29
TIR (Thermal)	–	B10, B11	10 – 12

how DEMs integrate with SWAT model sub-basins

 

Digital Elevation Model (DEM) is used to divide a watershed into multiple sub-basins in the SWAT (Soil and Water Assessment Tool) model. Each colored rectangle represents a sub-basin automatically delineated based on terrain and stream network derived from the DEM.

Here's how the integration works in SWAT:

• The DEM provides elevation values that define surface flow direction and accumulation.

• SWAT uses this to identify outlet points and delineate sub-basins based on topographic divides.

• Each sub-basin is further subdivided into Hydrologic Response Units (HRUs) using land use, soil, and slope data.

This process is covered step-by-step in the course: 👉 ArcSWAT Model with ArcGIS - Run for Any Study Area – GIS

Applications of DEMs in GIS

Digital Elevation Models (DEMs) are indispensable in hydrology-focused GIS applications, where understanding the movement of water across landscapes is critical for managing natural resources, predicting flood risks, and planning sustainable infrastructure. Among their most powerful uses is the delineation of watersheds, which represent the area of land that drains to a common outlet, such as a stream or river. Watershed boundaries are derived from the natural flow of surface water, which can only be accurately modeled when detailed elevation data is available. Using a DEM, GIS software can calculate flow direction by analyzing the steepest descent between neighboring pixels. This forms the basis for further computations such as flow accumulation, stream network generation, and ultimately, watershed delineation. These steps are essential in hydrological studies, as they allow for identifying upstream and downstream catchment areas, modeling water volume and speed, and simulating the spatial dynamics of surface runoff.

In practical applications, this capability is foundational to hydrological modeling frameworks like the Soil and Water Assessment Tool (SWAT), which relies on DEMs to break a watershed into sub-basins and Hydrologic Response Units (HRUs). A well-prepared DEM ensures accurate slope, aspect, and stream connectivity values—all of which influence surface water movement, sediment transport, and nutrient loading in SWAT models. Learners and professionals seeking to master this process from data preparation to simulation can explore the course ArcSWAT Model with ArcGIS - Run for Any Study Area – GIS. This course teaches how to use DEMs for watershed delineation, HRU creation, and full SWAT model execution in ArcGIS, making it highly valuable for hydrologists, water planners, and environmental researchers.

Beyond model simulation, DEM-based watershed analysis has wide-reaching impacts. It helps identify critical source areas (CSAs) where runoff, erosion, or pollution loads originate—guiding interventions in soil conservation, agriculture, and stormwater management. In flood modeling, DEMs support terrain-based inundation mapping, allowing analysts to simulate how far floodwaters may reach based on surface gradients. DEMs also aid in drainage network analysis, helping determine the natural or altered pathways of water, which is crucial in both rural and urban planning. Additionally, watershed-scale terrain analysis is essential for groundwater recharge studies, as topography influences infiltration zones and aquifer recharge potential.

For those looking to build strong foundational GIS skills—including DEM manipulation, stream extraction, and terrain modeling—using software like ArcGIS and ERDAS Imagine, the Complete Remote Sensing and GIS - ArcGIS – ERDAS course offers practical exposure. It covers elevation processing, slope analysis, and raster-based hydrological tools with real-world data, ideal for learners aiming to bridge remote sensing concepts with watershed applications. In sum, DEMs are not just digital surfaces—they are the backbone of watershed science and hydrological modeling in GIS, driving both analytical precision and environmental decision-making across scales.

Historical Background of DEM

 The concept and development of Digital Elevation Models (DEMs) have evolved significantly over the past century, driven by advances in cartography, remote sensing, photogrammetry, and computing power. Initially, terrain elevation was represented through topographic contour maps, meticulously created through field surveys and manual interpretation of aerial photographs. These early efforts provided critical elevation information but were time-consuming, labor-intensive, and lacked the flexibility and precision offered by digital models. The idea of transforming analog elevation data into digital raster format first gained momentum in the mid-20th century, as geographic and military organizations began to digitize terrain data for analytical modeling. One of the earliest known digital elevation models was developed by the U.S. Geological Survey (USGS) in the 1970s, where elevation data was stored in grid format and used for terrain visualization and basic topographic analysis.

The 1990s saw a surge in DEM development with the advent of satellite remote sensing, enabling elevation capture at global scales. However, a true turning point came in 2000 with the launch of the Shuttle Radar Topography Mission (SRTM)—a joint project by NASA, the National Geospatial-Intelligence Agency (NGA), and the German and Italian space agencies. This mission used radar interferometry onboard the Space Shuttle Endeavour to scan the Earth’s surface and produce what became the first nearly-global DEM, covering most of the land surface between 60°N and 56°S. The SRTM provided elevation data at 90-meter resolution globally, and later at 30-meter resolution for the U.S., revolutionizing the accessibility and quality of global terrain data. For the first time, researchers, planners, and developers could download accurate, standardized elevation data for most parts of the world—an unprecedented milestone in the field of GIS.

Following SRTM, a range of missions contributed to the global DEM landscape. The ASTER GDEM (Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model) was launched by NASA and Japan’s METI, providing global elevation data at approximately 30-meter resolution using stereo imaging techniques. Another advancement came from the TanDEM-X and TerraSAR-X missions by the German Aerospace Center (DLR), which used radar interferometry to produce one of the most accurate and detailed global DEMs to date, with 12-meter resolution and 2-meter vertical accuracy. Meanwhile, airborne technologies such as LIDAR (Light Detection and Ranging) became widely adopted for high-resolution terrain modeling, especially in urban and forested environments.

Today, DEMs are an indispensable part of GIS and remote sensing workflows, embedded in everything from flood simulation models and transportation planning to mobile navigation apps and virtual globe platforms. The growing number of open-access DEM repositories and cloud-based GIS services has further democratized their use, making them accessible not only to governments and scientists but also to students, NGOs, and hobbyists. Understanding the historical evolution of DEMs provides valuable context for their current capabilities and limitations. For learners seeking to engage with DEMs using real data and modern tools like ArcGIS and ERDAS, the Complete Remote Sensing and GIS - ArcGIS – ERDAS course serves as an ideal starting point, combining historical insight with practical, hands-on experience in terrain analysis.

Pixel Value as Elevation

 In a Digital Elevation Model (DEM), the most fundamental yet powerful concept lies in how each pixel’s value directly represents an elevation measurement—typically in meters above mean sea level. Unlike regular raster images that display color or intensity values for visualization, a DEM is a scientific dataset where each cell (pixel) stores a real-world numeric value that corresponds to the terrain’s height at that specific geographic location. These pixel values are not arbitrary; they are the result of remote sensing methods like stereo photogrammetry, LIDAR, or radar interferometry, and they form the basis of all terrain analysis in GIS. When visualized in software like ArcGIS, QGIS, or ERDAS Imagine, a DEM may appear as a grayscale image, but behind each pixel is a measurable quantity that can be used for complex spatial modeling tasks such as watershed delineation, slope stability analysis, or even infrastructure design.

The spatial resolution of a DEM determines the size of each pixel, usually expressed in meters. For instance, a 30-meter resolution DEM (like that from SRTM) means each pixel covers a 30 × 30 meter square on the ground, and the elevation value within that pixel is a representative average or interpolated height for that square. Higher-resolution DEMs, such as 10-meter or even 1-meter LIDAR-derived models, provide finer detail and are especially useful in urban, hydrological, and engineering applications. However, it's essential to understand that a DEM does not reflect vertical features like buildings or trees unless it is a DSM (Digital Surface Model). In a typical DTM or ground-level DEM, the pixel value is meant to capture the bare-earth elevation.

The real strength of DEM pixel values lies in their computational utility. With each pixel tied to a specific location in a coordinate reference system, these elevation values can be processed using mathematical algorithms to derive secondary terrain attributes. For example, by comparing the elevation of neighboring pixels, software can calculate slope (degree of incline), aspect (direction of slope), and hillshade (simulated sunlight exposure). In hydrology, pixel values are used to compute flow direction, accumulation, and stream networks, which are essential for watershed analysis and flood modeling. In 3D modeling, these values are extruded to create digital terrain surfaces, enabling visualization from different angles and even fly-through animations.

Moreover, DEMs can be queried just like any other raster layer. Analysts can extract elevation profiles along a path, calculate statistics over a specific region (e.g., average or maximum elevation), and perform cut-and-fill volume estimations for civil engineering. These operations would not be possible if pixel values were symbolic or visual only. Because the elevation values are real, scientific, and standardized, DEMs can also be merged or compared across time to detect land subsidence, volcanic growth, or glacial retreat—making them invaluable in climate monitoring and environmental assessment.

For learners beginning their journey into GIS and terrain modeling, mastering how pixel values in a DEM translate into measurable elevation is a crucial step. The Complete Remote Sensing and GIS - ArcGIS – ERDAS course offers real-world examples and tutorials on loading DEMs, interpreting pixel values, and applying them in practical GIS projects, ensuring a solid foundation in raster-based elevation analysis.

Creation of DEMs Using Image Overlapping Techniques

 The creation of Digital Elevation Models (DEMs) using image overlapping techniques is rooted in the fundamental principles of stereoscopy and photogrammetry, where the spatial difference between two or more overlapping images is used to calculate elevation. This method is one of the most widely used and intuitive approaches to generating 3D surface data from 2D satellite or aerial imagery. The core idea is based on parallax—the apparent shift in the position of a surface feature when viewed from two different angles. When an area on the Earth’s surface is imaged multiple times from different positions, the displacement of features across the image pairs can be measured. These measurements are then converted into height values using geometric triangulation methods, forming a structured grid where each pixel represents a precise elevation point.

In practice, stereo pairs are generated by capturing two high-resolution images of the same location from different vantage points—often from two separate satellite passes or by a dual-camera system mounted on the same satellite (such as Cartosat-1 or SPOT). The processing software identifies matching features across the two images and computes depth by analyzing the horizontal disparity between them. Each match corresponds to a 3D point in space, and when this is done across the entire image, it results in a dense point cloud that represents the terrain. These point clouds are then interpolated to form continuous raster grids—the final DEMs.

The quality of DEMs created through overlapping image techniques depends on several factors, including the resolution of the input images, accuracy of the satellite's positioning and orientation data, the viewing angle (or baseline distance) between the image pairs, and the surface texture of the terrain. Rugged or textured surfaces like mountains and forests tend to produce more accurate DEMs due to the ease of feature matching. Conversely, flat or homogeneous areas such as deserts, snowfields, or water bodies often result in poor correlation, leading to data voids or inaccuracies.

Moreover, the use of multi-angle stereo triplets (three images taken at different angles) can significantly improve DEM accuracy by increasing redundancy and reducing the chances of mismatches. Advanced photogrammetric software automates much of this process but still requires human oversight to ensure proper image alignment, remove mismatches, and filter noisy data points. The final DEM undergoes post-processing steps such as gap filling, smoothing, and validation against ground control points (GCPs) or existing elevation datasets.

Overlapping image techniques are commonly used in aerial photogrammetry, but they are also essential in satellite missions such as ASTER GDEM and Cartosat DEM production, where image-based DEMs provide extensive elevation coverage. This approach is especially valuable for projects in inaccessible or remote regions where LIDAR or ground surveys are not feasible. The resulting DEMs are critical inputs for slope analysis, hydrological modeling, visibility analysis, and even land cover classification in GIS workflows. Complete Remote Sensing and GIS - ArcGIS – ERDAS course, where students work hands-on with real elevation datasets and learn how to visualize, process, and apply DEMs in terrain-based analysis projects.

How Satellites Capture DEM Data

 Satellites capture Digital Elevation Model (DEM) data through sophisticated remote sensing techniques that measure the height of the Earth’s surface from space. These measurements are not direct visual images but are derived from differences in viewing geometry, signal return times, or phase shifts—depending on the sensor technology involved. The two primary methods used by satellites to capture elevation data are stereoscopic imaging (photogrammetry) and radar interferometry (InSAR). Both are highly effective, but they function based on different principles and are suited to different types of terrain and accuracy requirements.

In stereoscopic imaging, the satellite captures at least two images of the same area from slightly different angles. This is typically done either by using a pair of cameras mounted with a known baseline (in the case of satellites like SPOT or Cartosat) or by capturing the same area at different times during a satellite's orbit. This method mimics human depth perception, where the distance between our eyes allows us to perceive depth by comparing differences in the two images. Using photogrammetric principles, software algorithms calculate the displacement of features (called parallax) between the two images and convert that into elevation data. This technique is especially effective over terrain with strong texture—like mountains or valleys—where parallax is easier to detect.

Another highly advanced technique is Synthetic Aperture Radar Interferometry (InSAR), which is used by missions such as the Shuttle Radar Topography Mission (SRTM) and TanDEM-X. In this method, radar waves are emitted from the satellite to the Earth’s surface and reflected back. When two radar images of the same location are captured from slightly different positions, the phase difference between the radar signals can be calculated. These phase shifts correspond to differences in elevation and are used to construct highly accurate 3D terrain models. Unlike optical methods, radar interferometry has the advantage of penetrating through clouds and capturing elevation data at night or during adverse weather, making it ideal for consistent global elevation mapping.

Some satellites, like ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer), also use stereo imaging from optical sensors mounted on a single platform to create global DEMs, particularly in mountainous and rugged regions. While not as precise as LIDAR or radar methods, these optical systems are still widely used due to their availability and moderate accuracy.

It's important to note that while satellites provide raw data, the processing and modeling of DEMs are just as crucial as their capture. Raw elevation data must undergo geometric correction, gap filling, noise reduction, and often interpolation to become usable as a DEM in GIS platforms. These final DEMs are usually delivered in raster formats such as GeoTIFF or IMG files and are georeferenced to global coordinate systems like WGS 84.

Overall, satellite-based DEM acquisition has enabled large-scale, consistent, and repeatable elevation mapping of nearly every landmass on Earth. These datasets are critical for applications in topographic modeling, climate change studies, and hazard risk assessment. Understanding how satellites capture this data is fundamental for any GIS professional working with elevation models. Learners can get hands-on experience working with DEMs derived from SRTM and other satellite missions in the Complete Remote Sensing and GIS - ArcGIS – ERDAS course, which includes practical exercises on visualizing and analyzing elevation data captured from space.

Types of DEMs

 Digital Elevation Models (DEMs) are categorized based on the features they represent and the way surface elevation is recorded. Although the term “DEM” is often used generically, it actually encompasses several distinct types of elevation datasets, each tailored for specific applications in geospatial analysis. The two most commonly referenced types under this umbrella are Digital Surface Models (DSMs) and Digital Terrain Models (DTMs), each with their own characteristics and implications for GIS-based projects. A Digital Surface Model (DSM) includes the elevation of everything on the Earth’s surface—this means not just the bare ground, but also trees, buildings, vehicles, and other man-made or natural features. DSMs are particularly useful for urban planning, telecommunications (e.g., line-of-sight analysis for tower placement), and 3D visualization, where the height of surface objects matters. In contrast, a Digital Terrain Model (DTM) attempts to capture the "bare earth" by filtering out surface obstructions like vegetation and structures, resulting in a cleaner representation of the underlying topography. DTMs are especially crucial for hydrological modeling, geomorphological studies, and infrastructure development, where accurate terrain contours and slopes must be analyzed without surface interference.

Beyond DSMs and DTMs, the core DEM—as it is most commonly used in GIS—is often understood to represent a DTM in raster format. However, some countries and agencies maintain stricter definitions. For instance, in the United States Geological Survey (USGS) context, a DEM refers specifically to a raster grid of elevation values, whereas a DTM may also include vector data such as breaklines or spot elevations. Understanding these nuances becomes essential when selecting datasets for analytical tasks. For example, a hydrological model requiring water flow routing will yield more accurate results with a DTM, while a 3D city model visualized for shadow analysis or drone flight path simulation would benefit from a DSM.

The choice between DSM, DTM, and generic DEM depends on both the data source and the intended use. Some satellite-derived datasets like the SRTM (Shuttle Radar Topography Mission) are closer to DTMs but may include some surface features in forested areas. On the other hand, LIDAR-derived DSMs are extremely detailed, capturing buildings, treetops, and even individual vehicles, and can be processed further to create high-resolution DTMs using filtering techniques. Elevation models can also vary by resolution—ranging from coarse 90-meter DEMs to ultra-fine 1-meter LIDAR-based DSMs. Higher-resolution models offer more detail but also demand more storage and processing power.

In GIS applications, correctly distinguishing between DSMs and DTMs is crucial to achieving valid results. Misusing a DSM where a DTM is required, or vice versa, can lead to analytical errors—such as misestimating watershed boundaries, flood zones, or visibility ranges. Therefore, users must not only understand what type of elevation model they are working with but also ensure it aligns with the objectives of their spatial analysis. For those new to this subject, hands-on exposure to different elevation models in software like ArcGIS and ERDAS Imagine is recommended. The Complete Remote Sensing and GIS - ArcGIS – ERDAS course offers a strong introduction to this topic, allowing learners to visualize, compare, and apply various DEM types in real GIS projects.

What is Digital Elevation Model (DEM)

 A Digital Elevation Model (DEM) is a digital representation of the Earth's surface topography, where elevation values are systematically recorded in a grid-based format. In this raster model, the Earth’s terrain is broken down into uniform cells (also known as pixels), and each cell stores a single value that represents the elevation at that specific location, usually in meters above mean sea level. Unlike traditional contour maps that represent elevation using lines and require interpretation, DEMs provide elevation data in a continuous, easily quantifiable form that is computationally efficient and highly suitable for spatial analysis. DEMs serve as the foundational dataset for a wide variety of applications in the geospatial sciences, ranging from hydrological modeling and flood risk assessment to landform classification and infrastructure planning. Since each pixel corresponds to a georeferenced location, DEMs can be layered with other geographic data, such as land use, soil, and rainfall, making them indispensable in GIS-based modeling workflows.

What distinguishes a DEM from other elevation datasets is its simplicity and versatility. It is strictly a representation of the bare earth terrain and excludes any man-made structures, vegetation, or water bodies (although water surfaces can sometimes be modeled depending on the source data). This is especially important in environmental analysis, where surface roughness and vegetation canopy could distort interpretations of true ground elevation. By offering a standardized elevation reference, DEMs help scientists and engineers simulate water flow, assess terrain ruggedness, delineate watersheds, and perform countless other terrain-based computations. The quality of a DEM is determined by its spatial resolution—the finer the resolution (e.g., 10m vs. 90m), the more accurate and detailed the terrain representation. Modern high-resolution DEMs can even capture subtle elevation differences in urban or agricultural landscapes.

DEMs are often confused with other elevation data formats such as DSMs (Digital Surface Models) and DTMs (Digital Terrain Models). While DSMs capture the height of everything on the surface, including trees and buildings, and DTMs may include contour or breakline data, DEMs are purely raster-based and represent the Earth's surface in its most natural form. Their utility lies not only in their straightforwardness but in their adaptability across platforms. From 2D elevation mapping to 3D terrain visualization and simulation, DEMs have become a core component of GIS technology. With the increasing availability of open-source elevation data from satellite missions like SRTM and ASTER GDEM, and the growing demand for terrain analysis in fields like disaster management, civil engineering, and environmental planning, the role of DEMs is only set to expand. For beginners in GIS, understanding what a DEM is and how it behaves within a GIS environment is fundamental. Mastery of DEM usage opens doors to more advanced spatial modeling and geostatistical techniques, forming the backbone of nearly all elevation-based spatial analysis. You can learn about DEM analysis here 

Application of Machine Learning in GIS

 The integration of machine learning into image classification workflows has significantly transformed how land use and land cover (LULC) analysis is performed within the field of GIS. Traditional classification methods—often reliant on manual interpretation or simplistic thresholding—are no longer sufficient to manage the complexity of high-resolution satellite imagery and the nuanced patterns of urban growth, vegetation change, and land transformation. Machine learning algorithms, particularly supervised classifiers like Support Vector Machines (SVM) and Random Forests, have emerged as powerful tools capable of handling multidimensional spectral data and producing high-accuracy classification results. In land use mapping, these algorithms can distinguish between spectrally similar classes—such as fallow land and built-up areas—with greater precision by learning from training samples and adapting to subtle differences in pixel behavior. However, successful implementation requires a thorough understanding of preprocessing steps, training sample strategy, classification parameters, and accuracy assessment. For learners and professionals looking to master this process within ArcGIS, the course Landuse Landcover with Machine Learning Using ArcGIS Only offers a comprehensive guide to performing end-to-end classification using Sentinel-2 imagery and SVM, without requiring external platforms. It emphasizes real-world application and project-based learning—perfect for those aiming to produce publishable or operational outputs. Furthermore, beyond current land use assessment, machine learning plays a vital role in predicting future land dynamics. With historical LULC data as input, advanced models like CA-Markov, supported by ML-based classification, allow users to simulate how urban expansion or agricultural conversion might unfold in coming decades. For those interested in forecasting such changes, the course Future Land Use with GIS - TerrSet - CA Markov – ArcGIS offers a specialized focus on integrating classified images into predictive models, bridging the gap between current land cover mapping and scenario-based planning. Together, these two courses equip learners with both the analytical depth and technical skill needed to not only classify satellite images accurately but also to apply those classifications in forecasting future land transformations—a critical need in sustainable development, environmental monitoring, and urban planning.