# Australian varying landcover in ACCESS-rAM3 (JULES+UM)

**URL:** <https://forum.access-hive.org.au/t/australian-varying-landcover-in-access-ram3-jules-um/6043>\
**Category:** Experiments\
**Created:** [12 March 2026 00:43 UTC](https://forum.access-hive.org.au/t/australian-varying-landcover-in-access-ram3-jules-um/6043 "2026-03-12T00:43:31Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![mlipson](https://avatars.discourse-cdn.com/v4/letter/m/5f8ce5/32.png) [@mlipson](https://forum.access-hive.org.au/u/mlipson)\
**Post date:** [12 March 2026 00:43 UTC](https://forum.access-hive.org.au/t/australian-varying-landcover-in-access-ram3-jules-um/6043/1 "2026-03-12T00:43:31Z")

</div>

**Experiment title:** Australian varying landcover in ACCESS-rAM3 (JULES+UM)

**Summary:** The Bureau of Meteorology Hydrological Modelling Team have developed two new landcover and LAI datasets from 24 years of remotely sensed data 2000-2023. One is a climatology over that period, the second is monthly varying. @siyuan has run this full period offline in JULES, comparing 1) IGBP, 2) CCI, 3) BoM climatology 4) BoM monthly varying.

We now wish to evaluate differences in coupled land-atmosphere runs to understand if feedbacks alter the effects seen offline. These experiments will also form part of the Land Working Group plans for assessment between JULES and CABLE.

**Scientific motivation:**

Most climate and weather models use climatological or static data to describe landcover and LAI. For example, for the datasets on Gadi used by ACCESS here: `/g/data/access/TIDS/UM/ancil/atmos/master/vegetation/cover`

- IGBP landcover @ 1 km is averaged data between [April 1992 through March 1993](https://www.usgs.gov/centers/eros/science/usgs-eros-archive-land-cover-products-global-land-cover-characterization-glcc?utm_source=chatgpt.com).
- CCI landcover @ 300 m is averaged between [2008-2012](https://www.esa-landcover-cci.org/?q=node/158) (the Bureau and Met Office use the 2010 epoch for NWP).

For Australia landcover and LAI can vary significantly between years because of relatively high variability in inter-annual rainfall. Static datasets will not capture this variation.

We are interested to understand the impact of using a more realistic time-varying landcover and LAI product, like the BoMs.

**Experiment Name:**???  
**People:** @mlipson @siyuan  
**Model:** ACCESS-rAM3  
**Configuration:** [GitHub - 21centuryweather/HM\_coupled: Hydrological Modelling (HM) experiments with coupled rAM · GitHub](https://github.com/21centuryweather/HM_coupled/tree/main)  
**Initial conditions:** BARRA-R2 (through ACCESS-rAM3)  
**Run plan:** Depends on available resources.

BoM Hydrological Modelling (ce10) have ~ 800 kSU available. Assuming 12 km resolution over Australia with GAL9 settings, this allows ~ 13 years of runtime. As we wish to compare:

- CCI
- BoM static
- BoM varying

This provides approx 4 years of simulation. If we are able to draw on LWG resources, this could be extended, which would also change the period we would like to run.

**Simulation details:**  
**Total KSUs required:** 0.8 - 1.5 MSU  
**Total storage required 🔔:**  
**Outputs:**  
**Restarts:**  
**Related articles:**

**Analysis:**

**Conclusion:**