Can Perennial Crops Help Agriculture Survive a Water-Constrained Future?

Agriculture's toughest question is now shifting from how to save water to how to redesign land use before scarcity forces the decision. A new study published in Agriculture examines this problem through the lens of perennial crops, the long-lived tree and vine systems that include nuts, citrus, grapes and stone fruits.

The research, titled "Optimising Regional Land Use to Enhance Water Productivity Under Climate Uncertainty: The Role of Perennial Crops," was authored by Karin Schiller, James Montgomery, Marcus Randall and Andrew Lewis, affiliated with Bond University, the University of Tasmania and Griffith University. It uses the Murrumbidgee Irrigation Area in New South Wales, Australia, as a case study, but its policy relevance extends far beyond one farming region. The authors note that the area shares features with other food-producing regions facing similar resource pressures, including California's Central Valley, the Mediterranean, the Horn of Africa and South Africa's Western Cape.

Perennial crops are both valuable and vulnerable. They can generate high economic returns, but they also require long-term water access and capital commitment. Unlike annual crops, which can be changed, rotated or skipped in drought years, perennial systems lock farmers into decisions that may last decades.

The paper notes that perennial systems can take three to seven years to generate economic benefit, while breeding programs can take three to fifteen years. This rigidity makes land-use planning a climate adaptation problem. If regions expand high-value orchards without understanding future water risk, they may improve short-term returns while increasing long-term exposure. If they avoid perennials entirely, they may miss opportunities to raise economic water productivity.

Perennials can lift returns, but they lock in risk

The research extends a digital decision-support model known as the spatio-temporal agricultural land use sequencer, or STALS, to better represent perennial crops. The model tests two climate projections: ACCESS1.3, described as hotter and drier, and CanESM2, described as warmer and wetter. It also includes five water-availability scenarios, soil-based land management units and deficit irrigation assumptions.

The study divides the 141,000-hectare case region into land parcels and tests different shares of land assigned to perennial crops: 0%, 25%, 37.5%, 50%, 75% and 100%. It evaluates outcomes across three planning periods: the 2020s, 2050s and 2090s.

The model predicted an economic capacity of about AUD 2 billion in the 2020s when 25% of land was assigned to perennial crops, a result the authors say aligns with observed farm-gate commodity values. More broadly, the results show that including perennial production systems helps maximise regional net revenue, although the economic benefit differs depending on future climate and water availability.

However, the study does not argue for a simple expansion of orchards and vines. Perennial crops can improve economic water productivity, yet they also increase exposure to long-term water, infrastructure and market risks. The paper warns that expansion should be evaluated against on-farm and regional infrastructure constraints, as well as establishment costs.

High-value crops may deliver more revenue per unit of water, but food systems cannot be planned only around price. Annual crops may generate lower farm-gate value, but they support domestic food security, staple supply and production flexibility. The study notes that annual crops such as horticulture, cereals and pulses remain important for food security and trade, while nuts are predominantly export-oriented.

Deficit irrigation turns scarcity into a planning tool

Water scarcity does not always eliminate perennial crops from the future landscape. The model shows that deficit irrigation can help keep some perennial systems feasible under constrained conditions.

Deficit irrigation involves applying less water than the crop's full requirement in a controlled way. In the study, deficit irrigation regimes were tied to water availability: drought conditions supplied 7% of crop water requirement, very low allocation supplied 20%, low allocation supplied 50%, and mid-high allocation supplied 100%.

The model treated annual crops as requiring full satisfaction of water needs, while perennials could remain feasible through deficit irrigation. Under hotter and drier conditions, deficit irrigation was activated more frequently, especially in the 2050s and 2090s. The authors conclude that deficit irrigation improved the feasibility of perennial crops, particularly under the hotter and drier ACCESS1.3 climate projection.

Water policy often focuses on allocation volumes, trading rules and infrastructure, but the study shows that operational decisions at farm level, how water is reduced, spread and timed, can reshape what land uses remain viable.

Deficit irrigation may preserve feasibility, but it can affect yield, quality and long-term plant health. The authors acknowledge that the model used classic deficit irrigation and did not include other approaches such as partial root-zone drying or regulated deficit irrigation. It also adjusted yield but not commodity quality.

Climate-smart farming needs portfolios, not silver bullets

Resilient farming regions will need crop portfolios, not single-crop solutions. Diversity matters because it spreads risk across production systems, markets and climate conditions.

The authors find that crop diversity was highest where annual and perennial systems coexisted. At the extremes, 0% perennial or 100% perennial, diversity declined because some crops were excluded from the model. This is a critical insight for policymakers tempted to push high-value export crops as a water-productivity solution. Maximising economic return is not the same as maximising resilience.

Soil-crop matching is equally important. The study shows that climate-smart landscapes depend on aligning crops with land management units, soil attributes and projected climate conditions. Figure 6, for example, shows that self-mulching clay can support a wider range of perennial options than deep sandy soil, where crop choices are more constrained. The authors state directly that optimising productivity requires land-use selection to be aligned with projected production environments.

Climate adaptation in agriculture cannot rely only on broad national crop strategies. It needs localised decision tools that combine water availability, soil type, crop biology, market demand, food security priorities and long-term climate uncertainty.

The study also identifies important limitations. Long-range optimisation is difficult to validate. The model does not include groundwater, grey water or ecosystem land services, and its market and yield adjustments are not commodity-specific. Future work should incorporate more complex deficit irrigation tactics, dynamic chill modelling, land-use transition costs, commodity-specific pricing and improved yield simulations under future climate conditions.

On the whole, climate-smart agriculture is not just about saving water. It is about deciding which crops belong where, under which climate future, with what level of risk, and for whose benefit. The future of farming in water-stressed regions will be shaped less by one breakthrough crop than by smarter combinations of crops, soils, water strategies and planning tools.