Geospatial Monitoring and Modeling System
Land Change Modeler TerrSet is an integrated feature-rich software
system for monitoring and modeling the Earth
system for sustainable development. Based
GeOSIRIS
on three decades of development within the
Graduate School of Geography at Clark University,
Habitat and Biodiversity Modeler TerrSet offers a constellation of integrated vertical
and horizontal applications.
Ecosystem Services Modeler
Earth Trends Modeler
Climate Change Adaptation Modeler
IDRISI GIS Analysis
IDRISI Image Processing
1
TerrSet Geospatial Monitoring and Modeling System
The Land Change Modeler (LCM) for analyzing land cover change, empirically temperature, precipitation, vegetation productivity and the like. ETM is an exceptional
modeling its relationship to explanatory variables and projecting future changes. LCM tool for the assessment of climate change in the recent past (e.g., the past 30 years).
also includes special tools for the assessment of REDD (Reducing Emissions from
Deforestation and forest Degradation) climate change mitigation strategies. The Climate Change Adaptation Modeler (CCAM, pronounced “see cam”)-a tool for
modeling future climate and assessing its impacts on sea level rise, crop suitability and
GeOSIRIS-a unique tool for national level REDD (Reducing Emissions from species distributions.
Deforestation and forest Degradation) planning, developed in close cooperation with
Conservation International. With GeOSIRIS, one can model the impact of various The IDRISI GIS Analysis Tools – a broad spectrum of fundamental tools for GIS analysis,
economic strategies on deforestation and carbon emissions reductions. primarily oriented to raster data. Special features of the IDRISI tool set include a suite
of multi-criteria and multi-objective decision procedures and a wide range of tools
The Habitat and Biodiversity Modeler (HBM) for habitat assessment, landscape for statistical, change and surface analysis. Special graphical modeling environments
pattern analysis and biodiversity modeling. HBM also contains special tools for species are also provided for dynamic modeling and decision support. IDRISI also provides
distribution modeling. a scripting environment and an extremely flexible application programming interface
(API) that allows the ability to control TerrSet using languages such as C++, Delphi
The Ecosystem Services Modeler (ESM) for assessing the value of various ecosystem and Python.
services such as water purification, crop pollination, wind and wave energy, and so on.
ESM is based closely on the InVEST toolset developed by the Natural Capital Project. The IDRISI Image Processing System – an extensive set of procedures for the restoration,
enhancement, transformation and classification of remotely sensed images. IDRISI has
The Earth Trends Modeler (ETM)-a tool for the analysis of time series of earth the broadest set of classification procedures in the industry including both hard and
observation imagery. With ETM, one can discover trends and recurrent patterns in soft classification procedures based on machine learning (such as neural networks) and
fundamental earth system variables such as sea surface temperature, atmospheric statistical characterization.
Land Change Modeler
Land cover change is one of the largest imprints of
humans on the Earth system. Understanding and
projecting transitions into the future is critical for smart
development, biodiversity conservation and managing
climate mitigation strategies such as REDD (Reducing
Emissions from Deforestation and forest Degradation).
The Land Change Modeler (LCM) analyzes historical
land cover data to assess, model and predict land cover
change using powerful machine learning procedures.
Major components of LCM include:
Land Change Analysis
Given two historical land cover layers, quickly generate
graphs and maps of land change, including gains and
losses, net change, persistence and a breakdown of
The Land Change Modeler (LCM) in TerrSet uses map layers of historical change in combination with layers of potential explanatory
contributors to each transition. LCM includes the ability
variables (such as slope, proximity to roads, and so on) to develop empirical models of change. From these, projections can be
to generalize complex transitions using trend surface made of the expected land cover at a future date. The LCM dialog to the left shows the process of specifying explanatory variables
analysis. associated with a specific transition and the use of the Multi-Layer Perceptron (MLP) neural network to learn their association with
the transition. The result is a series of transition potential maps (upper left). These are then used in combination with Markov Chain
Transition Potential Modeling analysis to generate a projected land cover map for a future date (lower-center). In addition, LCM can generate a soft projection—a
Modeling the potential of land to experience specific map of the vulnerability to land transition (upper-right).
transitions (such as deforestation for agricultural
development) lies at the very heart of LCM. Using
2
historical land cover layers along with a set of potential explanatory variables (such REDD Analysis
as proximity to roads, soil type and slopes), LCM uses empirical modeling tools to REDD—Reducing Emissions from Deforestation and forest Degradation—is a climate
establish the relationship between them. LCM currently supports logistic regression, change mitigation strategy that offers developing countries incentives to reduce their
a modified KNN (K-nearest neighbor) and an exceptionally powerful Multi-layer forest carbon emissions. Critical to the development of a REDD project is the ability
Perceptron (MLP) neural network for model development. In the latter case, a very to model expected land cover change in the absence of intervention. LCM is the ideal
detailed accounting is provided of model skill and the degree of contribution of each tool for this purpose and provides additional tools to quantify baseline carbon stock
explanatory variable. changes, including both CO2 and non-CO2 greenhouse gas emissions, and the possible
associated leakage from displaced baseline activities. LCM’s REDD tools follow the
Change Prediction World Bank’s BioCarbon Fund methodology and have been successfully employed in
Using transition potential models as a foundation, LCM uses Markov Chain analysis establishing approval for projects submitted to the Verified Carbon Standard program.
to project the expected quantity of change and a competitive land allocation model
to determine scenarios for a specified future date. Options exist to incorporate
planning interventions such as incentives and constraints, proposed reserve areas and
infrastructural changes.
GeOSIRIS
from LCM), biomass, soil carbon, peat, potential agricultural revenue, site accessibility
GeOSIRIS is a national-level REDD (Reducing Emissions from Deforestation and forest and administrative divisions (at multiple levels). Parameters include the global price of
Degradation) planning tool that quantifies and maps the impacts of REDD policies on carbon, the national reference level for emissions as a proportion of business-as-usual,
deforestation, carbon emissions, agricultural revenue, and carbon payments. Whereas national/sub-national proportional cost/benefit sharing, price elasticity of demand for
LCM’s tools are appropriate for the development and assessment of individual projects, agriculture on newly deforested land and exogenous increases in price due to changes in
GeOSIRIS is oriented to the evaluation of alternative economic policy decisions for demand. With these inputs, users can calculate the effective opportunity cost to establish
reducing greenhouse gas emissions and distributing REDD revenues. This application the relationship between deforestation and potential agricultural revenue. Following
is a spatial implementation of the OSIRIS economic model created by Conservation this, a wide variety of map outputs can be produced including expected carbon
International, the World Resources Institute and the Environmental Defense Fund. The emissions with and without the REDD policy in place, agricultural revenue, expected
model requires map layers of forest cover, forest cover change (which can be output deforestation, cost efficiency of emissions reductions, and so on.
GeOSIRIS is a national-level planning tool for REDD (Reducing Emissions from Deforestation and forest Degradation). Developed in partnership with Conservation
International, GeOSIRIS allows the user to specify the price of carbon, national level objectives for carbon emissions, the distribution of offset income between national
and local levels, and a variety of economic parameters such as price elasticity of demand for agriculture on newly deforested land. GeOSIRIS empirically determines the
effective opportunity cost based on a set of environmental variables and a provided map of potential agricultural revenue. Outputs include maps of expected carbon
emissions with and without the proposed REDD policy as in the example for Indonesia above.
3
Habitat and Biodiversity Modeler
The Habitat and Biodiversity Modeler (HBM) is a toolset designed for Conservation GIS professionals. It provides tools for the modeling of species distributions, habitat assessment,
habitat change and gap analysis, biodiversity analysis, and the planning of reserves and biological corridors.
Species
At the species level, HBM provides a wide variety of
options for the modeling of species distributions based
on observed locations and bioclimatic variables. Options
are provided for presence only data (Mahalanobis
Typicality and MaxEnt), presence/absence (Logistic
Regression and a Multi-Layer Perceptron neural
network) and abundance data (Multiple Regression). For
individual species, the Habitat Assessment tool allows
the mapping of primary and secondary habitat as well
as potential corridor areas. Facilities are also provided
for the assessment of habitat change and gaps in the
protection system.
Biodiversity Analysis
HBM’s Biodiversity tab provides a special utility for
working with the IUCN (International Union for
Conservation of Nature and Natural Resources) Red The Habitat and Biodiversity Modeler (HBM) includes a wide range of tools for habitat assessment and species distribution/
List of Threatened Species database. Subsets of species biodiversity modeling. In this illustration, the IUCN Red List species database for terrestrial mammals (freely available on line) was
ranges can be extracted based on the Red List threat scanned to extract the range polygons of species endemic to South America. Once extracted, the polygons were then used to create
level, location and endemicity. Given a set of extracted a map of frequency using the alpha diversity mapping option.
ranges (typically hundreds to thousands of ranges),
maps of alpha diversity, gamma diversity, beta diversity,
Sorensen’s dissimilarity, and range restriction can be available, the process of change can be mapped. For example, a landscape may not be particularly fragmented, but
generated. may be actively fragmenting. Other processes that can be detected include deformation, perforation, shift, shrinkage,
enlargement, attrition, aggregation, creation and dissection.
Landscape Analysis
At the landscape level, a facility is provided for landscape Planning
pattern analysis that maps normalized entropy, relative For planning, tools are provided for biological corridor analysis and reserve planning. The corridor tool allows you to
richness, edge density, patch area and patch compactness. specify the target width and number of branches as well as produce maps of development suitability and conservation
In cases where land cover maps from two dates are value. It then maps corridors of least biological risk. For reserve planning, we have provided an integrated interface to
the well-known Marxan reserve planning tool.
Ecosystem Services Modeler
The Ecosystem Services Modeler is a spatial decision support system for assessing the • Water Purification: calculates nutrient retention and the avoided cost of
value of natural capital for sustainable development. The Ecosystem Services Modeler water treatment.
provides 15 ecosystem service models that are based closely on the InVEST toolset • Sediment Retention: estimates sediment retention in a watershed and the avoid-
developed by the Natural Capital Project. Fully integrated into the TerrSet system, users ed cost of sediment removal.
can access a powerful platform for deriving the data for use in the 15 models, such as • Carbon Storage and Sequestration: estimates the amount of carbon currently
deriving future land use scenarios using the Land Change Modeler and predicted future stored or sequestered in the landscape and its value.
precipitation using the Climate Change Adaptation Modeler to assess the impacts on • Timber Harvest: models total biomass and net present value of harvested timber
water yield or sedimentation. at the parcel level.
• Habitat Quality and Rarity: assesses the impacts of anthropogenic threats on the
• Water Yield: measures the average annual yield based on inputs such as precipitation, quality and rarity of habitats.
land cover, evapotranspiration. • Crop Pollination: evaluates the abundance and economic value of pollinators for
• Hydropower: estimates the potential energy production from a reservoir and agriculture.
its value. • Habitat Risk Assessment: an alternative to the Habitat Quality and Rarity approach,
4
the Habitat Risk Assessment (HRA) model focuses on
management strategies and exposure/consequences.
• Offshore Wind Energy: evaluates potential wind en-
ergy sites and their economic value.
• Aesthetic Quality: models the visibility/visual impact
of developments. The TerrSet approach includes a full
consideration of atmospheric conditions on visibility.
• Overlapping Use: calculates the frequency and impor-
tance of human activities within a management zone.
• Coastal Vulnerability: evaluates the exposure of
coastal communities to storm-induced erosion and
inundation.
• Marine Aquaculture: estimates the productivity and
economic value of existing aquaculture systems or
identifies the best potential sites for new aquaculture
projects.
• Wave Energy: evaluates wave-generated energy provi-
sion potential including consideration of the siting of
The Ecosystem Services Modeler (ESM) is closely based on the InVEST tools developed by the Natural Capital Project. Fifteen
wave energy conversion facilities. ecosystem service models have been implemented. In this illustration, ESM is used to evaluate water yield in the year 2100. The
predicted land cover input was generated by LCM and the predicted precipitation was generated by CCAM.
Earth Trends Modeler
Over the past 30 years there has been a veritable explosion
of earth observation image time series such as monthly
global sea surface temperature and vegetation index
imagery. The Earth Trends Modeler (ETM) is a ground-
breaking tool for the analysis of such data. With ETM, one
can map long term trends, trends in seasonality (such as
vegetation phenology) and search for recurrent patterns
over time, over space or over space-time. Additionally,
ETM’s pattern seeking tools are particularly adept at
looking for coupled patterns across multiple image series.
For example, the patterns associated with ENSO (the El
Nino Southern Oscillation) in oceanic, tropospheric and
stratospheric temperatures are related, but different in
space and lagged in time. ETM has the tools to discover
such patterns and much more, including:
The Earth Trends Modeler (ETM) is specially designed for the analysis of earth observation image time series. In this illustration, a
series of 348 global images of monthly NDVI vegetation index data were analyzed for the presence of trends in seasonality. Pixels
• Interactive tools for the exploration of image time se-
colored gray (which are almost absent) indicate a stable seasonality. All other colors represent trends. ETM provides an interactive
ries, including animation of space time cubes, tem- legend (lower left) to interpret the trend for any area chosen either by a vector boundary or a circular area of interest (the area in
poral profiling and the graphing of seasonal trends eastern Alabama and western Georgia in this case). The green curve represents the beginning of the series (1982) while the red one
for user-defined regions of interest. Profiles can be represents the end (2010). The X axis is time and the Y axis is NDVI. As can be seen, NDVI has generally increased with a growing
generalized by superimposing a linear trend, a Theil- bimodality. Spring is coming a bit earlier (11 days) and the autumn is extending longer (almost 30 days).
Sen median trend, a polynomial trend (up to 9th or-
der), a moving average, a Gaussian moving average
or a moving maximum.
5
• Mapping of long-term trends including degree of linearity, degree of non-linear • Spectral analysis, including Fourier-PCA and Wavelet Analysis (Maximum
monotonicity, rate of change (both parametric OLS and non-parametric Theil- Overlap DWT).
Sen median trend), trends in seasonality, and trend significance for both linear • Linear modeling with image series dependent variables and multiple image or in-
and seasonal trends. dex series dependent variables. Dependent variables can also be lagged forwards
• Space-time pattern analysis. Techniques include Principal Components (Em- or backwards in time. Outputs are in the form of images of correlation, partial
pirical Orthogonal Function analysis), Extended PCA, Empirical Orthogonal correlation, R2 and adjusted R2. Residual series can also be output.
Teleconnection (EOT) analysis, Extended EOT, Multichannel Singular Spectrum • Preprocessing tools include gap filling (missing data interpolation), deseasoning
analysis (MSSA) and Canonical Correlation analysis (CCA). CCA and the ex- and denoising. Tools are also provided for detrending and pre-whitening (trend-
tended modes allow the analysis of multiple series. Further, in most cases, the preserving removal of serial correlation).
user can select either S-mode to focus on patterns over time, or T-mode to ex-
tract patterns over space.
Climate Change Adaptation
• Generate global temperature and precipitation climate scenarios using any one,
Modeler or an ensemble, of 20 atmosphere/ocean general circulation models (AOGCMs)
based on user-selectable greenhouse gas and sulfur dioxide emission scenarios.
The Climate Change Adaptation Modeler (CCAM – pronounced “see-cam”) addresses • Derive up to 19 bioclimatic variables (such as maximum temperature of the
the growing challenge of adapting to a rapidly changing climate. The Climate Change warmest month, minimum precipitation of the driest month, etc.) for use in spe-
Adaptation Modeler is a suite of tools for modeling future climate and assessing its cies distribution models (using HBM).
impacts on sea level rise, crop suitability and species distributions. It includes tools • Model the impact of future projections of temperature and precipitation on crop
for climate scenario generation using the National Center for Atmospheric Research’s suitability. Crop parameters can be added for any crop, but the included database
MAGICC and SCENGEN models (based on CMIP3/AR4), crop suitability modeling provides the parameters for almost 2000 crops.
using the Food and Agriculture’s EcoCrop database, modeling sea level rise impact, • Model the effect of projected sea level rise incorporating both the uncertainty of
downscaling climate projections, and the derivation of bioclimatic variables. With the projection and the uncertainty in the elevation model.
CCAM, you can, for example:
The Climate Change Adaptation Modeler includes an interface to the National Center for Atmospheric Research’s MAGICC and SCENGEN models for generating
climate scenarios (based on CMIP3/AR4). In this illustration, the MAGICC suite of coupled gas-cycle, climate and ice-melt models was used to determine changes in
greenhouse-gas concentrations and global-mean surface air temperature (the graph) using the A1B-AIM scenario. SCENGEN was then used to generate an ensemble
forecast for January 2100 based on 20 atmosphere/ocean general circulation models.
6
IDRISI GIS Analysis
The foundation of TerrSet is the IDRISI GIS Analysis and Image Processing system. All components use the IDRISI API and the IDRISI data file structures. While IDRISI was once
a standalone product, it is now incorporated within the TerrSet System.
Like all GIS software systems, IDRISI handles both vector and raster data. However, its orientation is primarily raster. While vector systems are focused on the management of
objects in space, raster systems focus on space itself. Using a fine grid data structure, raster systems have exceptional analytical power. With their powerful implicit topology, raster
systems are excellent for modeling geographic processes, particularly those that involve flows over space.
Map Algebra / Modeling
Perhaps the most fundamental property of raster systems
is their ability to treat map layers as variables in an
equation. TerrSet’s IDRISI GIS provides an extremely
rich set of logical and mathematical operators such as
one would find on a scientific calculator. In this manner,
mathematical models such as the Universal Soil Loss
equation can be solved. The basic tools such as arithmetic
operators, exponentiation, trigonometric functions and
logarithmic operators can be performed either through
the use of simple dialogs, or written as equations using
the Image Calculator. In addition, TerrSet provides a
very powerful graphical modeling tool complete with
feedback loops and iterations.
Distance and Context Operators
The IDRISI GIS component in TerrSet provides a very
The foundation for TerrSet is the IDRISI GIS suite of analytical tools. This figure illustrates some of the many GIS analysis tools
rich set of tools for the assessment of distance across
provided. These include a risk mapping based on a multi-criteria evaluation, a modeling of surface runoff incorporating information
space including Euclidian distance, non-Euclidian cost
on precipitation and soil infiltration and the use of the Image Calculator for basic map algebra tasks. The bottom right shows a view
distance where the effect of frictions to movement are
of the FlyThrough interactive 3-D flight viewer.
accommodated, to anisotropic cost distance where the
frictions are different in various directions. Tools are also
provided for optimal path analysis and spatial allocation Spatial Statistics
based on distance. Context operators (also known as IDRISI provides a wide range of tools for statistical analysis of map layers including descriptive statistics, point
neighborhood or focal operators) derive values based distribution measures, autocorrelation analysis, pattern and texture measures, polynomial trend surface analysis, linear
on individual cells and their neighbors. The IDRISI GIS and multiple linear regression, logistic regression and multinomial logistic regression.
includes a wide selection for filtering, pattern analysis and
determining rates and directionality of change (such as Decision Support and Uncertainty Management
slope and aspect). The latter can be used to describe force A hallmark of TerrSet’s IDRISI GIS has been the development of tools for multi-criteria/multi-objective decision
vectors and tools are provided for their combination to support and uncertainty management. This development continues, and now TerrSet incorporates a major graphical
derive resultant forces. modeling tool for multi-criteria and multi-objective decision support – the Spatial Decision Modeler (SDM). SDM
incorporates tools for use of fuzzy sets to convert variables into comparable factors, the Analytical Hierarchy process
Surface Analysis for the derivation of factor weights, Ordered Weighted Averaging for multi-criteria evaluation and a newly designed
Surface interpolation procedures in TerrSet’s IDRISI procedure for multi-objective land allocation (MOLA). The new MOLA allows one to set targets for allocations based
GIS include inverse distance weighting, triangulated on either area or accumulated cost of land acquisition. In addition, it allows one to control contiguity, the number of
irregular network (TIN) modeling, Thiessen polygons, contiguous allocations and the relative degree of compactness (specified as the minimum spanning distance). This is a
trend surface mapping and Kriging. Tools are provided first-of-its-kind implementation in the GIS industry and a major breakthrough in decision support.
to derive topographic features such as slope gradients,
aspect, illumination (hillshading), and curvature TerrSet’s IDRISI GIS also contains a wealth of additional procedures for uncertainty management including error
and facilities to delineate watersheds and viewsheds, propagation tools through Monte Carlo Simulation, the evaluation of decision risk as a result of propagated error,
determine surface runoff and flow patterns, evaluate calculation and aggregation of Fuzzy Sets, and the aggregation of indirect evidence to support a weight-of-evidence
sedimentation and model soil erosion. conclusion using both Bayesian and Dempster-Shafer approaches.
7
IDRISI Image
Processing
The final component of the TerrSet constellation is the
IDRISI Image Processing toolset. The raster foundation
of TerrSet provides an exceptional base for the analysis
of remotely sensed imagery. Tools can be broadly
classified into restoration, enhancement, transformation
and classification. However, while the system provides
a broad-spectrum approach to the analysis of remotely
sensed images, the emphasis is unquestionably on the
conversion of remotely sensed images to interpreted
maps. The tools for classification are the broadest in
the industry, including both hard and soft classification
procedures based on machine learning (such as neural
networks) and statistical characterization. Elements TerrSet includes the IDRISI Image Processing toolset with the most extensive set of classifiers for remotely sensed imagery in the
include: industry. In this example, a Landsat 8 image of Rhode Island for May 2103 was classified using a Maximum Likelihood classifier
and training data for classes consistent with the U.S. National Land Cover Database (NLCD).
Image Restoration
A complete set of geometric and radiometric correction
tools including resampling, mosaicking, destriping and Image Transformation
atmospheric correction. A special procedure is provided TerrSet’s broad set of raster tools provides unlimited possibilities for transformation. However, special facilities exist for
for Landsat import (including Landsat 8). Principal Components and Canonical Correlation Analysis and vegetation index creation.
Image Enhancement Image Classification
Image enhancement tools include stretching, TerrSet contains the broadest set of classifiers (procedures for the computer-assisted interpretation) for remotely sensed
compositing, filtering and sharpening images. For imagery in the industry. Supervised classifiers include parallelepiped, minimum distance, maximum likelihood, Fisher
systems that include a higher-resolution panchromatic Linear Discriminant Analysis, and K-nearest neighbor (KNN). Unsupervised classifiers include ISODATA, chain
band, pan-sharpening procedures include a modified clustering, histogram peak and K-Means. Machine-learning classifiers include Classification Tree Analysis (CTA)
IHS (Intensity-Hue-Saturation) or Hyperspherical Color and four neural network classifiers: Multi-Layer Perceptron (MLP), Self-Organizing Map (SOM), Fuzzy ARTMAP,
Sharpening (HCS) procedure. Spatial Fourier Analysis and Radial Basis Function (RBF). Soft classifiers and mixture analysis tools include a Bayesian probability classifier,
provides special opportunities for the removal of noise. a Mahalanobis distance classifier, fuzzy set classifier, and linear spectral unmixing. Hyperspectral image analysis and
classification is included. A special set of procedures are provided for segment-based classification. Tools for accuracy
assessment are also provided.
System Requirements CONTACT US
• Windows 7 and above, or Windows Server 2003 and above.
• Microsoft ACE 2010 or Microsoft Office 2010 or later.
• 1.3 GB hard drive space for application.
Clark Labs, Clark University
• 7.4 GB for Tutorial data. 950 Main Street
• 4 GB RAM, 8 GB or more recommended. Worcester, MA
• HD display (1920 x 1080) or greater recommended. 01610-1477 USA
TEL +1.508.793.7526
For more information, visit [Link]
FAX +1.508.793.8842
EMAIL clarklabs@[Link]
WEB [Link]