Map the local food-water system and its boundary

Local food-water resilience concerns the continuity and adaptation of essential functions during disruption, not merely the survival of one farm's output. Start with an observable public function: households in a defined area can obtain available and affordable food while the water services needed for basic use and production operate within known constraints. Household, farm, utility and market are different units of analysis. FAO city-region food-system guidance describes actor and process links that are useful for tracing local continuity pathways. It does not establish the facts of any particular city. A system map is a decision aid, not a claim that every arrow is a measured causal effect.

Set the boundary in three decisions. First specify space and time: two neighborhoods served by a wholesale market and water utility during a hypothetical eight-week dry season. Second include functions needed to answer the question: water supply and distribution, irrigation for local suppliers, food processing, transport, sale and household access. Third record what stays outside and how it enters: food sourced beyond the study area is an inflow at the market gate; fuel prices are an external pressure; provincial allocation decisions are governance constraints outside the neighborhood team's authority. Excluding a component does not make it irrelevant. It defines it as an input or constraint instead of an operational node. A question about pipe safety would require a different boundary and qualified specialists.

Draw named nodes and directed links. Nodes might include a utility, licensed well, farm, packing site, cold store, carrier, market and household group. A physical link needs a quantity, origin, destination and time period. Money links describe purchases or cost; information links carry alerts or decisions. A utility-to-market arrow is meaningless unless it names a flow, such as cleaning water supplied to a packing site. Label each link's evidence and confidence. Utility records may document distributed water, whereas interviews with three traders cannot establish a market-wide share. Treat a missing reading as unknown, not zero. FAO's city-region food-system guidance treats actors and processes as a connected network; the local map here remains a simplified training model and must be checked with affected actors.

Test the map with one disruption question: if water stops for two days, which function is exposed first and where does missing evidence stop the inference? Describe timing, not an unconditional causal arrow. With one day of stored water, the washing site is exposed on day two; a suitable replacement arriving after ten hours would change that conclusion. Do not add market stock to weekly deliveries as though both shared a denominator: stock is a quantity at an instant, whereas deliveries are a flow over seven days. Ask who may activate an alternative, who verifies its suitability and whose water allocation might be displaced. These governance and equity questions are not answered by a physical flow diagram. Check the boundary with growers, market staff, utility staff and household representatives before using it to allocate scarce resources.

Before taking the map to a decision committee, make a compact record for every arrow: relationship, unit, period, data owner, last update and possible failure point. 'Washing site to market: 40 saleable boxes daily, from five days of delivery logs' is different from 'market to households: number of households served unknown.' Do not convert boxes into households with an undocumented multiplier. Keep the water-to-washing relationship separate from the washing-to-food-availability relationship; observing the first cannot quantify the second. External arrivals or stock may compensate, while demand itself may change. Mark each step as measured, assumed or awaiting verification. Ask which missing fact would change today's contingency decision. Here it is the usable stock, continuity of external arrivals or duration of washing interruption.

Boundary review is not a ceremonial endorsement meeting. Ask the washing operator for the minimum operating water supply, the trader for safe reserve life, a household delegate for the difficulty of alternative access, and the utility for restoration authority. They may disagree about facts; retain each account with its source instead of averaging away the conflict. If residents identify another supplier absent from market logs, record a possible link for verification rather than treating it as either proven or irrelevant. Put one decision question to the committee: which missing fact, if checked today, would change the box-allocation decision? In this case it is reserve condition, continuity of incoming ready boxes, or outage duration. That question focuses data collection and keeps the map from becoming an endless inventory.