iHARP

What iHARP does

iHARP is an NSF-funded, university-affiliated research institute headquartered at the University of Maryland, Baltimore County (UMBC) in Catonsville, Maryland. It was established in 2021 as one of five NSF Harnessing the Data Revolution (HDR) Institutes under a five-year, $13 million grant (NSF Award #2118285). The institute integrates data science, machine learning, and physics-informed modeling with polar science to study ice sheet dynamics, sea level rise, and climate change impacts in the Arctic and Antarctic. It operates as a multi-institutional consortium currently spanning nine partner universities plus collaborators including NASA JPL, USRA, NVIDIA, IBM, Amazon Web Services, and the Texas Advanced Computing Center.

iHARP's technical portfolio centers on physics-informed machine learning and geospatial AI for polar applications. Named proprietary frameworks include the Icebed Mapping Application (DeepTopoNet-based deep learning with physics-guided residuals for subglacial bed topography from sparse radar), CMAD (unsupervised convolution matrix anomaly detection for 2D spatio-temporal data, applied to Antarctic sea ice), MT-IceNet (multi-temporal deep learning for Arctic sea ice forecasting, with reported 60% reduction in prediction error), and spatial-link plus iharp-query-executor published as Python packages on PyPI. Outputs are distributed through open science channels: GitHub repositories, Zenodo datasets, the iHARP Toolkits Portal, and the annual PolDS ACM SIGSPATIAL workshop series.

The institute does not operate a commercial revenue model. All operations are funded through the NSF HDR grant and supplemented by partner in-kind contributions (HPC from TACC, cloud from AWS, equipment/partnerships from NVIDIA and IBM). "Customers" in the conventional sense are academic researchers, students, federal agencies, and the broader public who consume its open datasets, tools, and publications. The institute has approximately 10 core staff with engagement from 80+ stakeholders and scholars across the consortium.

Name

iHARP

Legal name

NSF HDR Institute for Harnessing Data and Model Revolution in the Polar Regions

Website

https://iharp.umbc.edu

Company type

Private

Founded year

2021

Operating status

Operating

Headcount range

1–10 employees

Short description

iHARP is an NSF-funded research institute at UMBC that uses physics-informed machine learning and geospatial AI to model polar ice sheet dynamics, sea level rise, and climate change impacts for academic researchers, federal agencies, and the public.

Ownership category

akta.pro rank

Product category

Scientific Research Software for Polar and Climate Data Science

NAICS

Geophysical Surveying and Mapping Services (54136), Geophysical Surveying and Mapping Services (541360)

akta.pro primary industry

Downscaled Climate Projections & Scenario Analytics (CMIP, SSP/RCP, Regionalization) (EUABALAB)

akta.pro secondary industries

Climate & GHG/Emissions Modeling & Scenario Analysis (Inventories, Projections, Net-Zero Pathways) (EUAHAKAH), Remote Sensing & Geospatial MRV (Satellite, Aerial, LiDAR, SAR) (EUABACAF), Climate & Emissions Data Management (Data Lake, Integration, APIs) (EUABAEAC)

Keywords

• Polar data science
• Geospatial analytics software
• Ice sheet modeling
• Climate research tools
• Physics-informed machine learning

Headquarters

HQ city

Catonsville

HQ country

United States

HQ region

North America

Offices1 record

Markets served

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Infrastructure, Operations, Marketing or Sales

Revenue model

1. NSF HDR Grant Funding: iHARP is supported by a five-year, $13 million grant from the National Science Foundation's Harnessing the Data Revolution (HDR) Big Idea program. NSF Award #2118285.

Go-to-market motion1 record

Distribution channels4 records

Marketing channels6 records

Core offering

iHARP develops and distributes open-source software tools, interactive web applications, and Python packages for polar data science and climate change research, focusing on physics-informed machine learning approaches to ice sheet dynamics, sea-level rise prediction, and geospatial analysis. Core offerings include the iHARP Toolkits Portal hosting Icebed Mapping and CMAD applications, plus Python libraries (spatial-link, cmad, iharp-query-executor) on PyPI, all built to support academic researchers, government agencies, and policy makers studying polar climate impacts.

Product overview

iHARP is an NSF-funded research institute that develops a portfolio of software tools and platforms for polar data science, geospatial AI, and physics-informed modeling. The core offering is the iHARP Toolkits Portal providing interactive web applications (Icebed Mapping, CMAD) and Python packages (spatial-link, cmad, iharp-query-executor) published on PyPI. Supporting offerings include educational programs (Talk Tuesdays, WiseByte Wednesdays workshops), the annual PolDS ACM SIGSPATIAL workshop series, the NSF HDR ML Challenge competition, an open science data catalog, and open access to research through GitHub, Zenodo, and Ghub platforms.

Differentiator

Problem solved

Functional benefit

Products and services

• iHARP Toolkits Portal Web portal providing access to interactive software, web applications, and open-source Python packages developed by iHARP for polar data science, geospatial AI, and physics-informed modeling.
• Icebed Mapping Application Interactive web application that visualizes predicted subglacial bed topography beneath the Greenland Ice Sheet using deep learning with physics-guided residuals (DeepTopoNet) trained on sparse ice-penetrating radar data, supporting ice flow and ice dynamics analysis for polar researchers.
• CMAD (Convolution Matrix Anomaly Detection) Interactive Application Interactive web tool implementing the CMAD framework for Antarctic anomalous melting detection, an unsupervised anomaly detection system for 2D spatio-temporal gridded datasets that surfaces spatially coherent clusters of extreme negative events without requiring labeled training data.
• spatial-link (Python Package) Python framework for detecting statistically significant linkages from spatio-temporal data, published on PyPI for use by researchers analyzing polar and geospatial datasets.
• cmad (Python Package) Lightweight Python framework for extreme spatio-temporal anomaly detection featuring adaptive IQR-based thresholding, published on PyPI for researchers performing unsupervised anomaly detection on 2D time-series data.
• iharp-query-executor (Python Package) Python library for querying and executing spatial data operations against iHARP data infrastructure, leveraging xarray, shapely, and geopandas for structured geospatial data processing.
• iHARP Data Catalog Curated collection of datasets relevant to polar science research conducted at iHARP, providing researchers with access to integrated polar observation datasets for climate and ice sheet analysis.
• iHARP Open Science Infrastructure Comprehensive open science infrastructure encompassing GitHub repositories for open models, Zenodo community collection for open data, and the Ghub platform for Open Polar Science community engagement.

Quantifiable outcome

• Up to 60% decrease in prediction error for Arctic sea ice forecasting with MT-IceNet model
• +1 more outcomes

Segments3 records

Ideal customer profiles2 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability8 records

Feature4 records

Partnerships

19 partnerships are on record, tiered core and major.

• NASA Jet Propulsion LaboratorycoreStrategic or Co-development PartnerResearch collaboration on polar science, glaciology, and climate modeling. Co-PI Dr. Mathieu Morlighem worked with NASA JPL for his PhD.
• Texas Advanced Computing Center (TACC)coreTechnology or IntegrationProvides HPC infrastructure including Frontera GPU CPU hybrid cluster, Corral data storage, and computing resources for iHARP research.
• NVIDIAmajorStrategic or Co-development PartnerIndustry partner supporting iHARP's AI and machine learning research for polar science applications.
• IBMmajorStrategic or Co-development PartnerIndustry partner collaborating on data science and AI research for climate and polar science.
• Amazon (AWS)coreTechnology or IntegrationAWS collaboration providing cloud computing resources. AWS Foundations and Geospatial Immersion Day workshops organized.
• Dartmouth CollegecoreStrategic or Co-development PartnerPartner institution led by Dr. Mathieu Morlighem (Professor of Earth System Science). Contributes expertise in ice sheet modeling.
• University of Colorado BouldercoreStrategic or Co-development PartnerPartner institution led by Dr. Aneesh Subramanian. Contributes atmospheric science and climate modeling expertise.
• University of MinnesotacoreStrategic or Co-development PartnerPartner institution led by Dr. Shashi Shekhar and Dr. Mohamed Mokbel. Contributes geospatial computing and spatial data systems expertise.
• University of Alaska FairbanksmajorStrategic or Co-development PartnerPartner institution contributing expertise in Arctic research, glaciology, and cryospheric sciences.
• Amherst CollegemajorStrategic or Co-development PartnerPartner institution contributing to polar science and climate research collaboration.
• University of North TexasmajorStrategic or Co-development PartnerPartner institution with Dr. Sahara Ali and VR/AR research collaboration including Virtual Ice Museum.
• University of Texas at AustinmajorStrategic or Co-development PartnerPartner institution contributing to data science and climate research through TACC collaboration.
• University of MarylandmajorStrategic or Co-development PartnerPartner institution in the University System of Maryland contributing to research collaboration.
• Towson UniversitymajorStrategic or Co-development PartnerPartner institution with Dr. Chhaya Kulkarni joining as faculty. Contributes to spatiotemporal data mining research.
• Georgia Institute of TechnologymajorStrategic or Co-development PartnerPartner institution contributing to technology and engineering expertise for polar science applications.
• University of Maryland Center for Environmental SciencemajorStrategic or Co-development PartnerPartner institution contributing environmental science expertise and CGC-SCIPE collaboration.
• NASA Universities Space Research AssociationmajorStrategic or Co-development PartnerResearch collaboration on space-based observations and satellite data for polar science.
• Imageomics InstitutemajorStrategic or Co-development PartnerHDR ecosystem partner collaborating on NextGen Day workshop and interdisciplinary research.
• A3D3 InstitutemajorStrategic or Co-development PartnerHDR ecosystem partner co-leading the 2nd FAIR HDR ML Challenge with iHARP at FARR Workshop.

Scale indicators4 records

Recent moves7 records

Expansion highlights6 records

Broad incumbents

• National Center for Atmospheric Research (NCAR): Federally funded atmospheric and climate science research center operated by UCAR. Comparable as a much larger, established institution producing climate projections and Earth system models that overlap directly with iHARP's ice sheet and sea-level rise forecasting focus.
• British Antarctic Survey: UK national polar research organization conducting Antarctic and Arctic science including ice sheet and sea-level research. Comparable as an established, government-funded polar science organization with overlapping research programs in the same physical domains iHARP studies.
• Lamont-Doherty Earth Observatory: Columbia University research center with extensive climate, oceanographic, and cryospheric science programs. Comparable as a much larger, well-established academic research entity producing climate observations and analyses that overlap with iHARP's sea-level and polar climate focus.
• Scripps Institution of Oceanography: Major UC San Diego research institution with deep expertise in glaciology, atmospheric science, and sea-level rise. Comparable as a long-tenured, large academic research entity producing climate and cryospheric observations that overlap with iHARP's polar focus.
• Goddard Institute for Space Studies (NASA GISS): NASA laboratory conducting climate and polar research including ice sheet modeling. Comparable as an established public-sector research entity with overlapping Earth system and cryospheric modeling capabilities and a long publication track record in the same domains as iHARP.

Emerging players

• BlueDot (formerly Aiscreen): Climate risk and infectious disease analytics company offering geospatial climate and health risk products. Comparable as a commercial geospatial analytics provider applying Earth observation and ML to climate-related risk, representing the type of private-sector path iHARP's methods could inform.
• Jupiter Intelligence: Commercial climate risk analytics firm serving insurers, asset managers, and governments with physical climate risk models. Comparable as a developer of climate risk and projection analytics that operates at the intersection of climate science and applied risk modeling, the commercial counterpart to iHARP's open research outputs.

Direct peers

• Imageomics Institute: Sister NSF HDR Institute focused on data-driven biological research using ML on organismal imagery. Closely comparable as a peer HDR-funded center with a similar open science, multi-institutional consortium model and explicit NSF program alignment.
• A3D3 Institute: Sister NSF HDR Institute advancing real-time AI for astrophysics and accelerator physics. Comparable as a peer HDR center funded under the same NSF Big Idea, co-hosting challenges with iHARP and operating an analogous multi-university research consortium.
• National Snow and Ice Data Center (NSIDC): Federally funded data center at CU Boulder that distributes cryospheric and polar datasets and tools. Directly comparable as a community-supported hub for polar data infrastructure and research, serving overlapping Arctic/Antarctic science user communities.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks6 records

Key highlights7 records

Customer concentration

• GitHubgithub.com/iharp-institute
• YouTubeyoutube.com/@nsfiHARP/videos
• LinkedInlinkedin.com/company/iharp

Revenue estimate

Valuation estimate

Number of profiles

Profiles2 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about iHARP