Methodology and Guidelines for Climate Change Impact Assessment of Agricultural Infrastructure Projects

A research service commissioned by the Korea Rural Community Corporation (KRC) to establish the methodology and guidelines for assessing climate change impacts on agricultural infrastructure improvement projects. The work was carried out as a joint contract, and the Integrated Watershed Management Institute was responsible for building the SSP-based future climate change scenarios.

Item	Detail
Client	Korea Rural Community Corporation (KRC)
Period	26 April 2023 – 23 December 2023
Contract type	Joint contract (Institute share 30%)
Institute’s scope	Building SSP-based future climate change scenarios

What we did

Station-based future weather database

For 76 weather stations, daily downscaled data were produced with the Simple Quantile Mapping (SQM) method across 4 SSP scenarios, 18 GCMs and 6 weather variables (precipitation, maximum temperature, minimum temperature, relative humidity, wind speed and solar radiation).

• For stations without solar radiation measurements, values were taken from the nearest of the 20 measuring stations using Thiessen polygons. Earlier downscaling work produced no solar radiation data for such stations
• Sunshine duration, evaporation and FAO Penman-Monteith reference evapotranspiration (ETo) were derived in addition
• The database was built to the format of K-HAS, the modelling system KRC uses in practice, so that it can be applied directly in design work that accounts for climate change

1 km grid-based future weather database

From the 1 km resolution downscaled data produced by the Rural Development Administration, daily area-averaged databases were built for 511 agricultural water districts and 167 municipalities. Reference evapotranspiration at 1 km resolution was produced with the same FAO Penman-Monteith method. Grid data suit forested areas and places where elevation strongly affects weather variables, such as Jeju Island, better than station-based data, and serve as the climate exposure input for climate change vulnerability assessment.

Reproducibility assessment and future projection

Past reproducibility and future projections were evaluated on the multi-model ensemble (MME) of 18 GCMs. Future periods were divided into 30-year spans counted back from 2100: near future (2011–2040), mid future (2041–2070) and far future (2071–2100).

Extreme climate characteristics were analysed by administrative district and by grid using the extreme climate indices of the ETCCDI (Expert Team on Climate Change Detection and Indices).

Results

• In the reproducibility assessment of the 1 km grid data over 30 years (1981–2010), the percent difference averaged over the Korean Peninsula was -1.74% for annual precipitation and -0.001% for mean temperature
• Precipitation, maximum temperature and minimum temperature all increase from the historical period towards the future periods, with larger increases under SSP5-8.5 than under SSP1-2.6
• Frost days (FD) decrease more towards the far future under the same scenario, and more under SSP5-8.5 within the same period, with coastal areas decreasing more than inland areas
• Temperature and reference evapotranspiration increase most along the coast. Annual precipitation increases most around the southern coast and Mt. Jiri, and maximum 1-day precipitation increases towards the far future and under SSP5-8.5

Why it matters

With a single climate model, the size and direction of projected change depend on the characteristics of that model. This work took the 18-GCM ensemble as its basis and presented that uncertainty alongside the projections, then delivered the results in the K-HAS format and water-district units KRC already uses — a form that can be applied directly in design and vulnerability assessment work.