1601 East-West Road, Honolulu, HI 96848 info@pacificrisa.org 808.944.7111

Building the Scenarios

Impacts of Future Climate on Groundwater Resources and Management in Maui, Hawaii.

The Pacific RISA used qualitative input from stakeholders and available spatial data sources to build a series of maps depicting potential future land cover for the island of Maui, which can be modified as needed to suit the needs of different users and projects.

Overview of the Data

The Pacific RISA generated four future scenarios for this project, using many different datasets to represent future land cover classes for the island of Maui. This page is intended as in introduction to the scenarios and associated maps. Below are descriptions of each scenario narrative, and a summary of the steps that were taken to build them.

The current version of the data provided on this page (Version 1.1) was finalized in October 2018. A previous version (Version 1) was finalized in May 2015 and has been has been updated to reflect the cessation of sugarcane farming activities in central Maui, based on a new land cover map representing circa 2017 conditions. These data were used as the input land cover classes for the future climate/hydrological modeling component of this project. The data is available for download and modification to suit the needs of different users and projects. The Pacific RISA assumes no responsibility for any modifications to or products derived from the original datasets. To download the ArcGIS shapefile that contains the most recent Version 1.1 of the future scenario land cover classes and associated source information, please click here. For a more detailed description of how the scenario maps were created, please read the metadata, which is attached to the downloadable shapefile in xml format. The previous Version 1 and associated metadata can be found hereIf you have any questions regarding the use of this data, please contact Laura Brewington, the Pacific RISA Program Manager, or check out the Frequently Asked Questions at the bottom of this page.

Building the Four Future Scenarios

The Pacific RISA generated four stakeholder-defined future scenarios for this project to engage decision makers in the research and increase the relevance of the scenarios to policy and public audiences. During numerous meetings that were held between 2012 and 2014, several main land use components emerged: forest conservation, urban development, agriculture, ranching, and streamflow restoration. These components formed the backbone of four narratives that described different “futures,” or menus of realistic management options for the island of Maui.

A set of 25 land cover classes was chosen to correspond to known parameter values for the hydrological model. Each future scenario narrative and key assumptions/components were translated into the 25 land cover classes. If participants did not agree with the spatial representation of a scenario assumption or component, the group identified potential alternative inputs and the maps were adjusted accordingly. Additionally, concepts evolved as each narrative component was depicted in a map. Four final scenario maps were produced.

Future Scenario I: Low development, high forest conservation, local agriculture

This scenario reflected an interest to see how climate change could impact groundwater recharge in a future where more emphasis is placed on maintaining existing and actively restoring native forest, preventing spread of invasive species, and adding diversified agriculture and biofuels for food and energy self-sufficiency. Almost all of the current native forest cover is retained, and high investments in watershed protection facilitate the construction of additional ungulate fencing and active forest restoration within fence boundaries in the high elevations or east and west Maui. Alien forest expands in unprotected or unmanaged areas, particularly on southeast Maui, but not within ranch boundaries. Biofuel production replaces some sugarcane in central Maui as a renewable energy source, while land owned by the Department of Hawaiian Homelands (DHHL) in west Maui is also designated for future biofuel production. Remaining area that was formerly sugarcane is fallow or grassland.

Diversified agriculture as a sustainable local food source expands on new IALs, and assumes streamflow diversions or reservoir catchment for irrigation. Grasslands expand in ranching areas to promote mob grazing techniques, and some ranchland with suitable regions is converted to diversified agriculture. The Hawaiian cultural practice of taro cultivation is promoted, and assumes additional surface water use. Urban development proceeds at a slow pace, with around 60% of the projects in the 2013 Maui Island Plan database achieving completion.

Variables in the dataset that correspond to this scenario: LC_F1, iluF1, Source_F1

Future Scenario II: Business-as-usual

This scenario is a status quo continuation of existing practices, reflecting what many stakeholders perceive to be Maui’s current trajectory. Most of the existing native forest is maintained, and some funding is allocated to create a limited number of new ungulate fences at the highest elevations, but no active forest restoration is done within them. Alien forest expands outside of fenced areas, including ranchlands and higher elevation areas that were not forested in 2010. Sugarcane production has ceased in the central Maui plain and is now fallow, with limited flow restored to currently diverted streams and the addition of supplementary groundwater pumping for irrigation. The current production of boutique crops like coffee and pineapple continue, along with existing diversified agricultural practices. Minor biofuel production is added to DHHL lands in west Maui, with undeveloped areas that are currently fallow shifting to ranching grasslands. Managed tree plantations are added to ranchlands, but some are invaded by alien forest expansion. All of the fully entitled development projects in the 2013 Maui Island Plan are built out, replacing almost all taro cultivation with impervious surfaces.

Variables in the dataset that correspond to this scenario: LC_F2, iluF2, Source_F2

Future Scenario III: Aggressive development, high energy production, low forest conservation

In this scenario, development proceeds aggressively, and environmental concerns are not actively addressed. Existing native forest shrinks in response to stalled funding for watershed conservation and no new ungulate fences are constructed, while alien forest expands to the limits of predicted suitable habitat, even at the highest elevations of west Maui. Biofuel production to meet rising energy demands and petroleum costs replaces sugarcane in the existing IALs in central Maui, now classified as “diversified agriculture,” while the surrounding sugarcane land becomes fallow. Ranchlands are extensively invaded by alien forest, particularly in south and east Maui, and grasslands are lost reflecting reduced livestock production. In response to population growth and commercial and housing demands for tourism, all development projects in the Maui Island Plan are completed, and build-out to the full limits of the designated future growth boundaries occurs.

Variables in the dataset that correspond to this scenario: LC_F3, iluF3, Source_F3

Future Scenario IV: Balanced development and forest conservation

The final future scenario stemmed from stakeholders’ interest in exploring groundwater resources under a future with balanced development, economic, and conservation interests. To facilitate comparisons across scenarios, this scenario was built using components of the other three future scenarios. Native and alien forest cover was selected from the first, conservation-oriented scenario, with added tree plantations, grasslands, and west Maui biofuel from the second, business-as-usual scenario. The extensive urban land cover, in conjunction with biofuel production on IALs in central Maui, was chosen from the third, development-oriented scenario. The addition of this scenario resolved a recurring tension throughout the scenario planning process – namely, that there was an assumed conflict between conservation and development.

Variables in the dataset that correspond to this scenario: LC_F4, iluF4, Source_F4

Frequently Asked Questions

Who created these maps?

  • The Pacific RISA generated the maps using input from interviews and workshops with people on Maui and Oahu who are concerned about future water resources and were responsible for management or regulation of natural resources on Maui. Study participants developed different feasible narratives about future land cover on Maui, and spatial information was collected or created to represent those different futures.

What are the source files that were used to build these maps?

  • There is a long list of data files that went into these maps to make them as plausible as possible. Some of that data is available online and listed in the metadata. Other datasets are not publicly available due to the sensitive nature of the information. Study participants reviewed the maps as they were being generated to ensure that they accurately reflected their ideas of future land cover and management options.

How should I use the dataset?

  • First, download and unzip the files and check that you have a GIS software package installed that’s compatible with ArcGIS shapefile formats. For those who do not have ArcGIS installed, QGIS is an excellent open source alternative with a user-friendly interface. Once you open the dataset, you’ll notice that there are over 500,000 polygons, and depending on your computer it may be slow to load. You can switch between land cover classes easily by changing the symbology from variable to variable.

What do the different variable names mean?

  • This is only a brief summary; please read the metadata for a complete description of each of the variables and their individual values.
  • POLYGON_ID: Unique identification number of each water-budget subarea
  • AREA_SQ_M: Area of each polygon, in square meters
  • LC_2017: Name of land cover or vegetation, 2017
  • LC_FX (example LC_F3): Name of land cover or vegetation, future scenario X
  • iluFX: Identification number of future scenario X land cover classes
  • Source_FX: Source(s) of derived future land cover for scenario X

Why are there so many polygons?

  • The number of polygons corresponds to the USGS water-budget model, described here, which predicts groundwater recharge based on the land cover and climate information inputs. We chose to join our land cover classes to the USGS polygons, because without it the geometry of our file was an arbitrary mashup of the many different input datasets. If you’re interested in the individual water budget components, the USGS report and associated data files can be downloaded here, and the POLYGON_ID fields in both shapefiles can be joined to assign water budget components to each of the scenario land cover classes.

You mention biofuel in the summaries of the scenarios, but there is no “biofuel” land cover class in the dataset. Why is this?

  • There are many different crop varieties that could be used as “biofuel,” ranging from sugarcane to hemp. Because of the diversity of options and uncertainty about the different hydrological components for each one, the land cover class “diversified agriculture” is used to represent all types of biofuel.

What do the “fog” and “no fog” distinctions mean in the land cover class names?

  • This designation was given by the USGS in their baseline 2017 land cover file that was used to make these maps, and refers only to forest and tree species. “Fog” simply means that the land cover is above the 2,000-ft elevation contour, the base of the fog-interception zone. “No fog” means that the land cover is below the 2,000-ft elevation contour.

When I zoom in on my study area or region of interest, some of the future land cover classes don’t make sense or look strange. Why is this?

  • As described in the metadata, future land cover classes were created by layering existing datasets on current and future types of land cover, resulting in the intersection of many different data sources. It would have been impossible to examine each of the 300,000+ polygons for “rationality” – instead, we chose not to modify the resulting maps and leave it to the user to make changes where they think they should be made.

Can I change the land cover classes or add and delete some information?

  • We encourage you to modify the shapefiles in any way you wish! Please remember, however, that our objective was to provide a set of agreed-upon, plausible future land cover maps for Maui, so any changes to the land cover classes should be carefully considered with your particular study participants and objectives in mind. The original files will remain available on our project website.

How should I cite these maps?

  • Please cite the most current version of the maps as follows: Brewington, L. 2018. Maui Future Land Cover Scenarios Version 1.1. East-West Center: Honolulu, Hawaii.