Highly Erodible Land Model (HEL)


HEL identifies land that is highly erodible. It identifies steep land that is at risk of mass movement. It identified land that should be under forest to reduce the erosion risk..


The Highly Erodible Land (HEL) system identifies land highly susceptible to mass-movement erosion using a combination of the erosion terrains derived from the NZLRI database, a 15‑metre digital elevation model (DEM) to determine topography, and land cover mapping. It considers the main forms of mass-movement erosion in New Zealand (landsliding, earth flows, and massive gullying).

Highly erodible land was defined as “land with the potential for severe erosion if it does not have protective woody vegetation” and was identified by:

  • defining slope thresholds for each erosion terrain and assigning all land above the threshold to HEL on the basis of landslide risk (thresholds ranged from 24° on weak Tertiary-age mudstone to 45° on hard greywacke)
  • assigning all pixels mapped with moderate or severe earth-flow erosion and gully erosion (from the NZLRI) to HEL
  • determining whether land identified as HEL has existing woody vegetation cover (if protected it is excluded from HEL). The HEL land was further classified in terms of whether it was or was not connected to a water course.

The procedure produces five classes of HEL:

  • high landslide risk – connected to stream
  • high landslide risk – not connected to stream
  • moderate earth-flow risk
  • severe earth-flow risk
  • gullying risk.

Because it uses the DEM to identify slopes, the HEL system is capable of higher spatial resolution than the NZLRI approach alone. At present the land cover map is derived from Ecosat (~2001), but this may be readily updated to 2008 using LUCAS. HEL was first piloted in the Manawatu–Whanganui region and forms the basis of their SLUI programme (https://www.mpi.govt.nz/funding-and-programmes/environment-and-natural-resources/hill-country-erosion-programme/). It was subsequently widened to include the whole of the North Island (Dymond et al, 2008), and now North Island regional councils use it for soil conservation planning. MAF also use it to evaluate regional council requests for soil conservation funds.

Latest Version Version 1
State of Development Released and final
Current Development Activity None

Development Contact

John Dymond
+64 6 353 4800
Landcare Research
Private Bag 11052
Manawatu Mail Centre
Palmerston North 4442,
New Zealand

Main Developers

  • Landcare Research


Outcome Areas Environmental
Management Domains Land, Natural Hazards, Freshwater
Subdomains Erosion/Sediment
Intended End Users Land managers, Policy makers
Spatial Dimensions 2D
Steady State or Dynamic Dynamic
Level of Integration Environmental

Input & Output Data

Key Input Data Land Cover Geology and soils. slope and elevation
Input Data Formats Raster-format (Imagine)
Key Output Data Erosion risk
Output Data Formats Raster-format (Imagine)


Open/Closed Source Open Source
Licence Type Data licence
Licence Cost
(Non Commercial)
Licence Cost
Licence Purchase Contact
John Dymond
+64 6 353 4800
Landcare Research
Private Bag 11052
Manawatu Mail Centre
Palmerston North 4442

User Information

Operating Systems MS Windows, Linux
Software Needed Imagine
User Interface Graphical desktop
Ease of Use Easy Just put in a land cover file and press run.
Use in Policy Process Plan (Policy Formulation), Do (Policy Implementation), Check (Policy Evaluation), Review (Issue Identification)

Dymond, John, Shepherd, James, Page, Mike (No date) Rollout of erosion models for regional councils. Report, Landcare Research, 37 pages.

Support No formal support, but you can ring John Dymond.
Users Forum No

Technical Considerations

Programming Language Imagine Spatial Modeller
Methods included for calibration and validation Synthesis of published literature.
Methods included for managing uncertainty Not Applicable. (At risk or not)
Analytical Techniques GIS
Keywords erosion risk, land sliding, earth flow, gully erosion, sediment, land cover
Linkages to other Models
Key References

Dymond JR, Ausseil A-G, Shepherd JD, Buettner L 2006. Validation of a region-wide landslide risk model. Geomorphology 74: 70–79. Dymond JR, Jessen MR, Lovell LR 1999. Computer simulation of shallow landsliding in New Zealand hill country. International Journal of Applied Earth Observation and Geoinformation 1(2): 122–131.

Associated Case Studies

HEL - Manawatu Whanganui Region

After the 2004 storm the Council is currently examining options to reduce hill country erosion risk and one of the first tasks is to (better) define and identify land that is highly erodible.