Method for evaluating flood plain risks

A computer method of evaluating risks associated with flooding, comprising the steps of:

• defining a first geographical region on the computer;
• inputting first data relating to the first geographical region into the computer, the first data including (i) a vertical distance of each grid cell to a relevant river grid cell, (ii) a horizontal distance of each grid cell to a relevant river grid cell, and (iii) a catchment area at each relevant river grid cell;
• developing a model to determine a probability of recurrence of flood-related events in the first geographical region on the computer for a defined recurrence interval by using a regression model that uses the characteristics (i), (ii) and (iii) as observed values in the regression model
• calibrating regression coefficients of the regression model by reference to an alternative source of information regarding an occurrence of flood-related events in the first geographical region;
• defining a second geographical region on the computer;
• inputting second data relating to the second geographical region into the computer, the second data including (i) a vertical distance of each grid cell to a relevant river grid cell, (ii) a horizontal distance of each grid cell to a relevant river grid cell, and (iii) a catchment area at each relevant river grid cell;
• and
• applying the regression model by using the second data as observed values in the regression model to determine a probability of occurrence of flood-related events by the computer in the second geographical region for the defined recurrence interval;
• wherein the regression model uses at least the following linear model,
• logit(p)=b0+b1VDIST+b2HDIST+b3AREA
• wherein p is the probability that the selected grid cell is flooded in the defined recurrence interval, VDIST is the vertical distance of each grid cell to the relevant river grid cell, HDIST is the horizontal distance of grid cell to the relevant river grid cell, AREA is the catchment area at the relevant river grid cell, and b0, b1, b2 and b3 are regression coefficients of the regression model that are calibrated based on a climate type of the respective geographic region.

A method of evaluating risks associated with flooding in a geographical region performed on a computer, comprising the steps of:

• defining an area to include a catchment area of a river flowing through the region on the computer;
• dividing at least a portion of the area into a plurality of grid cells;
• identifying data relating to characteristics of the area within one or more of the grid cells, the data including (i) a vertical distance of each grid cell to a relevant river grid cell, and (ii) a horizontal distance of each grid cell to a relevant river grid cell;
• and
• determining, by using a regression model that uses data from said identifying and data on the catchment area of the river as observed values, a probability of occurrence of a flood event by the computer in a selected grid cell for a defined recurrence interval;
• wherein the regression model uses at least the following linear model,
• logit(p)=b0+b1VDIST+b2HDIST+b3AREA
• wherein p is the probability that the selected grid cell is flooded in the defined recurrence interval, VDIST is the vertical distance of each grid cell to the relevant river grid cell, HDIST is the horizontal distance of grid cell to the relevant river grid cell, AREA is the catchment area at the relevant river grid cell, and b0, b1, b2 and b3 are regression coefficients of the regression model that are calibrated based on a climate type of the geographic region.

A method of evaluating risks associated with flooding in a geographical region performed on a computer, comprising the steps of:

• defining an area to include a catchment area of a river flowing through the region on the computer;
• dividing at least a portion of the area into a plurality of grid cells;
• identifying data relating to characteristics of the area within one or more of the grid cells, the data including (i) a vertical distance of each grid cell to a relevant river grid cell, and (ii) a horizontal distance of each grid cell to a relevant river grid cell;
• determining, by using a regression model that uses data (i) and (ii) from said identifying and data on the catchment area of the river as observed values, a probability of occurrence of a flood event by the computer in a selected grid cell for a first recurrence interval;
• and
• using a result of said determining to calculate a probability of occurrence of a flood event on the computer in the selected grid cell for a second recurrence interval;
• wherein the regression model uses at least the following linear model,
• logit(p)=b0+b1VDIST+b2HDIST+b3AREA
• wherein p is the probability that the selected grid cell is flooded in the defined recurrence interval, VDIST is the vertical distance of each grid cell to the relevant river grid cell, HDIST is the horizontal distance of grid cell to the relevant river grid cell, AREA is the catchment area at the relevant river grid cell, and b0, b1, b2 and b3 are regression coefficients of the regression model that are calibrated based on a climate type of the geographic region.