Historical, Current and Future Land Use Trajectories of the Kimberley Region, Western Australia

Executive Summary

The Kimberley region of Western Australia is renowned for its physical beauty, diverse biota, complex and relatively undeveloped coastline, and fascinating Aboriginal history. It also possesses immense resource potential. This potential has sparked a vigorous dialogue among government, industry, Aboriginal communities, and NGOs over how future development in the Kimberley can best achieve a sustainable future that effectively balances economic growth with environmental and social objectives, including preservation of the natural heritage of this relatively untouched region and the cultural values of its Aboriginal population.

The Kimberley region has substantial, though poorly quantified, reserves of fresh water, significant reserves of minerals and hydrocarbon fuels, and large tracts of land suitable for grazing and crop agriculture. Climate change may also treat the Kimberley more kindly than more southerly parts of Western Australia, thus increasing the value of economic development in the region. In addition, the Kimberley's proximity to Asian markets makes it an economically attractive location for export industries and tourist development.

However, the projected growth in economic activity from farming, ranching, mining, oil and gas, and tourism must inevitably bring with it increases in environmental impact. More people means more and larger settlements, increased road traffic (and, possibly, more roads), more shipping and air flights. More agriculture means altered hydrology and increased impact on natural ecosystems. An expanding resource-extraction industry means a larger footprint from mines, oil and gas facilities, and associated infrastructure. More tourism means more vehicle and boat traffic, as well as an increased demand for unspoiled natural areas. As terrestrial conditions change, this will have a direct impact on marine ecosystems through runoff and nutrient discharge, commercial and recreational fishing pressure, pollution from human developments, and other land use changes.

These changes will provide greater economic opportunities for residents of the Kimberley, particularly the Aboriginal peoples who make up about 45% of the region's population. Aboriginal peoples have extensive land and resource ownership rights, are responsible for cultural features across the landscape, and are central to the valuable land use of Aboriginal tourism. However, they also experience higher levels of chronic health problems, lower educational attainment, faster population growth, and higher mortality rates than the population as a whole. All regional stakeholders agree that any further economic development must realistically involve improving conditions for the region's Aboriginal populations.

Furthermore, the Kimberley's remoteness poses extra challenges regarding labour, logistics, and transportation, as well as trade with the rest of Australia. These challenges mean that any future economic growth in the region will need to cope with additional costs that may not be reflected in globally determined commodity prices. The task confronting those planning the future of the Kimberley is to devise future land use trajectories that maximize benefits and minimize economic, social, and environmental costs.

To address this issue, the Alces team of the Integrated Futures Program of Curtin University was contracted to conduct land use and landscape simulations in the Kimberley region. The project's goals were to describe the opportunities and challenges that are likely under different land use trajectories for the region, and to quantify how these changing terrestrial land uses may affect marine ecosystems.

The Alces approach explores land use trajectories from an integrated systems perspective that examines the cumulative effects of all major land uses upon a comprehensive set of economic, social, and environmental outcomes. The approach enumerates all the key land use sectors that are "in play" in the Kimberley and quantifies the current spatial footprint for each. It then asks when, where, and how these footprints are likely to change in the future, and attempts to assess how sensitive these predictions are to uncertainties in future growth trajectories.

No analysis, however detailed, can predict the future with any certainty, of course. Instead, the Alces models provide a way to explore future development scenarios to explicitly define the assumptions involved and reveal the kinds of changes that might occur, the key drivers of those changes, and the likely future consequences of current land use decisions. As such, they can represent a valuable aid to strategic planning deliberations.

To conduct these analyses, Alces has developed a customized, spatially explicit landscape/ landuse simulator, Kimberley Alces Online (KAO), to allow the Alces Group and its clients to explore past, present, and future relationships involving human land uses and natural disturbance regimes. KAO uses a map-based representation (5 x 5 m grid) of the Kimberley region layered with geospatial data on land uses (including residential, transportation, croplands, livestock, mining, oil and gas, forestry, and tourism/recreation), physical/climatic features (topography, soils, drainage, vegetation, temperature, precipitation, etc.) and natural disturbance (fire, insects, storms, etc.). Powerful calculators allow users to combine, filter, and display these data layers in a flexible, intuitive way that permits easy exploration of alternative scenarios.

A crucial step in this process is finding and evaluating the inputs to the models. The Alces models are only as good as the data that underlie them. Accordingly, the Alces team has spent considerable effort, in consultation with Australian federal, state, and regional govts, research institutes, resource companies, and other stakeholders to ensure that the information in our models is as good as possible. Nevertheless, in many cases data are not as comprehensive as we would like, and the Alces models are designed to accommodate better data as it becomes available. This report documents in detail the sources, assumptions, strengths and weaknesses of underlying data. Readers who need to understand the models at this level of detail should consult the appropriate sections of the report.

High-resolution maps of 450 of the most important data themes have been compiled into the Kimberley Biophysical and Land Use Atlas. This document, available online at https://www.dropbox.com/s/is3724plf9cazkd/Kimberley Atlas Show Version Oct 2 016.ppsx?dl=0, allows project stakeholders to familiarize themselves with the physical, biotic, and anthropogenic features of the region. If desired, it can also be used in schools as an educational tool.

The Kimberley region of northern Western Australia comprises 419,254 km2 on land and ~513,491 km2 in the marine domain. Its seasonal monsoonal climate features a warm wet season from November through March and a cooler dry season from April through October. However, potential evapotranspiration exceeds precipitation by a substantial margin even during the wettest months. Both temperature and precipitation vary dramatically both spatially and from year to year, leading to occasional seasonal extremes that can limit land uses more severely than long-term averages would suggest.

For much of the year, fresh water is relatively scarce in the Kimberley, and water resources are therefore important determinants of land use. Intense rainfall during the wet season recharges

streams and rivers, many of which dry down to disconnected pools during the dry season. As a result, the hydrological regime in the Kimberley is defined more by seasonal extremes than by annual averages. Damming, especially of the Ord River to create Lake Argyle, provides a reliable water source for irrigation near Kununurra. Groundwater reserves are incompletely documented, but some regions--especially the Canning Basin--show signs of aquifer depletion.

Intense rainfall and runoff events contribute to inherently high rates of soil and sediment transport in the Kimberley. Anthropogenic land uses--especially grazing, but also mining, crops, and transportation--have added significantly to this transport. This raises concerns about loss of soil fertility and altered coastal sediment dynamics, particularly in heavily affected areas. Damming of the Ord River has also likely caused a reduction in sediment transport from that river into the Timor Sea.

Fires have been an important contributor to the disturbance regime of the Kimberley region since prehistoric times. However, the nature of that disturbance has changed in the past few decades. Though the details remain a subject of debate, the general consensus is that fires have become fewer, larger, and of higher intensity than in the past. This creates a less heterogeneous landscape, leading to loss of biodiversity and increased soil erosion. KAO allows users to vary the parameters of current and past fires to explore the effect of shifting fire regimes.

The marine component of the Kimberley region is diverse, and its dynamics are incompletely documented. The KAO simulator allows users to simulate inputs of sediment and fresh water into the marine system and disturbances due to human land uses and population growth. Because most of the region's population, as well as most projected future population growth, is in coastal regions, this magnifies the effect of human activity on marine systems.

Climate change is likely to affect the Kimberley region in coming decades. Climate models suggest an overall warming, with average temperature increasing by up to 2.5 oC and extreme temperature events becoming more frequent. Changes to precipitation totals are likely to be relatively modest on average, with southern and western (coastal) parts of the region becoming drier and central and eastern parts receiving additional rain. However, climate models also forecast increased inter-annual variation, and this increased variability may override any effects on average precipitation levels.

Sea surface temperatures are likely to rise by 1 to 3 oC, contributing to an increase in incidences of coral bleaching. Ocean pH is likely to fall by .06 to .30 by 2090. Forecasts suggest that sea level will rise by ~0.5-1 m by 2100, which will cause a 6- to 30-fold increase in flood events.

Land uses within the Kimberley region currently include residential, conservation network, transportation, agriculture, aquaculture, mining, energy, tourism and recreation, and assorted other industrial activities. For each sector, KAO has projected future growth using low, medium, and high-growth assumptions to produce alternative "what-if" scenarios. Land-use planners should note that the vast majority of the Kimberley region lies within 20-40 km of Aboriginal cultural sites, and that it is important to incorporate this knowledge into land use decisions.

The current population of the Kimberley region is roughly 40,000 individuals, of which 40% is Aboriginal and 60% non-Aboriginal. Since the arrival of Europeans in the late 1800s, population has fluctuated considerably in response to changing land uses and economic development, particularly gold mining and pearling. Current government policy has the goal of increasing the regional population through increased economic development, and during the past few years the

population has grown at an average annual rate of 1.9% per year. Projections using low (1.5%/yr), medium (2.0%/yr) and high (2.5%/yr) population growth rates predict a regional population of ~80,000-~140,000 by 2064. This growth will be accompanied by comparable growth in residential area and demand for housing, water, food, and electricity.

Since the majority of infrastructure, population, and land use is located on the coast or along major rivers near the coast, impacts on marine ecosystems are to be expected. The Port of Broome, the largest centre for shipping, moved ~528M tonnes of goods per year in 2012-2013, an amount expected to increase to 956M tonnes by 2018-2019. In addition, recreational boat traffic accounts for an estimated 4.2M km of travel annually.

The agricultural crop industry in the Kimberley is in its early stages of development. So far, only a few hundred km2 are under cultivation, largely in the Ord River Irrigation Area, where ~270 km2 are now developed, with projections to grow to ~440 km2 within the next decade. The Ord River irrigation scheme currently supports over 60 different crop species, most of which is exported to markets in Asia. Future expansions are intended to focus on cotton, sugar, and chia production, with significant expansion of forage crops also under discussion.

The government of Western Australia expects that the food production system in the Kimberley will expand significantly in the future, to meet rapidly growing demand for food in and near the region. In particular, the possibility of an irrigated cropland sector in the La Grange region south of Broome is under active exploration. The KAO simulator is equipped to explore the economic and environmental impacts of this and other future trajectories of development.

Aquaculture activities in the region began in the late 1800s with the pearling industry. Pearling continues today, though the emphasis has shifted from pearl shell to cultured pearls. Pearl production is now at historic highs, although the workforce has declined an order of magnitude in the past century to a current level of ~200 workers.

One large barramundi aquaculture operation has been established at Cone Bay on the Buccaneer Archipelago, and expansions are under consideration. Though this sea-pen aquaculture uses minimal amounts of fresh water for processing and for the workforce itself, waste products can lead to local eutrophication of coastal waters, and attention to currents and waste dissipation will be an important consideration in planning.

The livestock industry in the Kimberley consists of typically ~600,000 cattle on ~23 million hectares of pastoral leases, with most animals being shipped away for finishing and slaughter outside the region. Developing a more fully integrated livestock production system would bring economic benefits, but is currently limited by shortages of dry season fodder, as well as the necessary infrastructure for finishing and slaughter. In addition, the drought-prone, semi-arid grasslands of the Kimberley are susceptible to degradation from high grazing pressure, particularly near water holes and along watercourses. Given the uncertainties that accompany climate change, some experts recommend careful, risk-averse management of stocking levels. The KAO simulator permits planners to explore the economic and environmental impacts of future grazing policies.

Mining leases in the Kimberley region exceed 70,000 km2, although the actual footprint of mines today is only ~70 km2, and much of this is not in active production. This reflects the mercurial nature of the mining industry, in which mines alternate between active and inactive status depending on commodity prices. Current gross water use by mines is estimated at ~36.5 M m3/yr.

One major new proposal is under development for a mineral sands mine in the Thunderbird Deposit, 75 km WSW of Derby.

The Kimberley region also contains significant hydrocarbon reserves, both onshore and offshore. However, development of these resources has been controversial, largely due to concerns about whether the economic benefits justify costs to environmental and Aboriginal values. A key task for KAO is to simulate future development of hydrocarbon reserves in the region, adjusting the pace and location of development according to multiple scenarios. Because the details of future development will depend on many uncertain variables (commodity prices, production costs, location of reserves, and policy decisions and public license to operate), these simulations should be regarded not as accurate predictions of the future, but as potential trajectories to explore the effect of changing assumptions.

The tourism industry attracts a total of ~3.7 million visitor-nights and $333M in visitor expenditures (2013 figures). However, the industry has experienced slow or stagnant growth in recent years, as competing locations such as Bali and Thailand have emerged as attractive alternatives. If the tourism industry is to grow at a rate of 2-3%/yr over the next several decades, as government targets specify, then existing tourism sites and infrastructure will need to grow comparably. Optimizing the geographic placement of this growth, and constructing a transportation network to bring tourists to those sites, is a key task confronting planners. Presently, most tourism activity and infrastructure occurs in or near Broome. Planners should explore the consequences of continuing that pattern vs pursuing a more broadly dispersed tourism strategy. KAO facilitates that exploration by mapping the location of current and potential tourist sites, including ocean cruise ship destinations, and the infrastructure that accompanies them.

One key component of the tourism industry is the Cape Leveque road through the Dampier Peninsula. This 200 km road is the single transportation artery up the peninsula, linking many of its most scenic sites. However, about half the road is unpaved, and this portion is technically demanding at best and not passable during extended periods of the wet season, making it a major impediment to the Peninsula achieving its full tourism potential. Proposals to pave the remaining portion of the road have received political support. If the work is completed, it is likely to increase travel by both residents and tourists between Broome and Aboriginal communities on the peninsula. This will increase tourism opportunities, but also increase demand for infrastructure and adverse effects on the environment.

Large portions of the Kimberley are recognized as part of a conservation reserve network, including national parks, marine parks, nature reserves, conservation parks, and indigenous protected areas. If planners take a landscape-level approach to coordinating these and any future conservation reserve set-asides, they can maintain environmental integrity during future development, and enhance the growing ecotourism industry in the region. KAO provides the tools to assess competing land uses, identify important potential conservation sites, and minimize conflict between these two goals.

For each of these land uses, KAO includes estimates of the economic benefits (jobs, gross revenue) that derive from that land use, and its contribution to gross regional product. This will allow planners to explore the tradeoffs in costs and benefits of policies that favor one land use over another.

Each of these proposed land uses, viewed in isolation, would seem to have a relatively small, and therefore tolerable, environmental impact. However, in the real world land uses do not occur in isolation. Instead, most realistic development trajectories involve changes in multiple land uses at the same time. To truly understand the likely consequences of development, planners must take into account the cumulative effects of all these changes.

Water management will be a critical consideration in any future development planning. Current water use in the Kimberley is ~427M m3/yr, with 82% of this being for irrigation, primarily in the Ord River region. Industrial (9%), rural (5%), domestic (3%), and commercial (1%) uses make up a much smaller fraction of total use. Water demand is likely to rise significantly as the region grows, with projected water demand of 897-1,469 M m3/yr by 2065. The magnitude of this increase makes it essential that planners quantify water availability from surface and groundwater sources, as well as addressing the levels of in-stream flow required to maintain ecological systems.

At present, the direct footprint of human development occupies only a tiny fraction (620 km2, or just 0.14%) of the total land area of the Kimberley region. Over the next 50 years, this footprint is likely to increase to between 1,300 km2 (2.1-fold increase, assuming low-range growth of 1.5% per year) and 2,130 km2 (3.4-fold increase, assuming high-range growth of 2.5% per year). Note that these estimates do not include the largest human land uses in the region, namely livestock grazing and conservation reserve.

Moreover, the effect of human land uses extends well beyond the actual footprint itself. Roads and other disturbances provide corridors through which invasive species can spread out into adjacent habitats; noise and commotion of human activity can cause animals to avoid otherwise suitable habitat in the vicinity of the footprint; changes in microclimate and hydrology can have effects felt well beyond the development itself. To account for this spreading effect, the Alces models allow users to specify a buffer zone around human land-use developments. Since different effects are likely to be felt at different distances, it is useful to examine buffers of several different sizes.

These buffers can greatly increase the extent of area affected by human activity. Adding just a 1 km buffer around anthropogenic features decreases the natural landscape from 99.9% of the Kimberley to just 72.2%. A 2 km buffer reduces natural landscape to 62.2%, a 5 km buffer to 40.9%, a 10 km buffer to 23.5%, and a 40 km buffer leaves just 0.2% of the total land area as undisturbed, natural landscape outside the buffer.

Moreover, land uses are not evenly distributed across the landscape. Instead, some areas experience a much heavier footprint than others, particularly in the Dampier peninsula, the Ord River and the Fitzroy region. In these more heavily impacted areas, landscape fragmentation can be a significant issue.

As human activity continues to increase in the Kimberley region, it poses increased risk of alteration of natural ecosystems and the services they provide. In particular,

Climate change is likely to alter temperature and precipitation throughout the region in largely unpredictable ways. This may lead to changes in natural disturbance regimes, especially fire and insect populations. These, in turn, are likely to trigger changes in plant communities and the wildlife that depends on them.

More human activity and increased grazing by livestock is likely to lead to greater runoff of sediment overland and in streams and rivers. This can reduce soil fertility, and the transport of sediment can alter the substrate of lakes, rivers, and estuaries and coastal ecosystems.

Increasing human footprint is already aiding in the establishment and spread of exotic plant and animal species. This is likely to become more of a problem in the future, exacerbated by possible degradation of grazing lands.

Increased diversion of surface and groundwater for human uses such as mining and agriculture, as well as to serve a growing human population, is likely to change the hydrology of the Kimberley region.

A growing human population will increase demand for recreational harvest of sport fish, especially near to settlements. If not carefully managed, this is likely to lead to changes in age structure of the fish population, which in turn will drive anglers--and their harvest pressure--to progressively more remote regions.

As climate change causes increased spatial and temporal variability in precipitation, livestock pressure on watering holes and other water sources is likely to become more extreme at times and places where droughts occur. This contributes to uneven grazing pressure, and especially to overgrazing along riparian rangelands, which is likely to lead to loss of grassland community structure, increased prevalence of invasive species, and loss of local biodiversity.

In contrast, climatic warming is likely to have relatively little impact on human land use, according to Alces simulations, because sufficient water reserves exist to mitigate the effect of increasing temperatures on crops and livestock. Natural variation in precipitation, watershed discharge, and sediment transport is already very high, so any increased variability due to climate change is likely to have only a small effect.

Future changes in human activity and climate change also present risks of alteration to marine environments of the Kimberley. In particular, climate change will alter the amount and variability of precipitation falling on the terrestrial portion of the region, which is likely to affect water and sediment discharge into the marine environment, in largely unpredictable ways. KAO landscape simulations may help explore some of this uncertainty.

Changes in population of coastal communities or along main-stem rivers can have a significant local effect on marine ecosystems, both through discharge of wastes and through other economic and recreational activities.

If offshore and nearshore hydrocarbon production increases in the coming decades, this will increase impacts due to pollution, increased human occupancy, boat traffic, and indirect effects such as increased fishing pressure.

In summary, the Kimberley region of northern Western Australia is at a development crossroads. Looking at the Kimberley today, it is easy to see it as an essentially undeveloped region full of "empty space", in which the human industrial footprint is so small that little effort need be devoted to managing environmental impact. In fact, however, the Kimberley region already experiences a wide-ranging network of cumulative effects of human activity. Current policies call for considerable economic and population growth in coming decades, which will only exacerbate those cumulative effects. In order to achieve maximum economic returns with minimum environmental costs, planners must be aware of cumulative environmental effects and include them in their discussions of future development.