Abstract. Crop growth in land surface models normally requires high-temporal-resolution climate data (3-hourly or 6-hourly), but such high-temporal-resolution climate data are not provided by many climate model simulations
due to expensive storage, which limits modeling choices if there is an
interest in a particular climate simulation that only saved monthly outputs.
The Community Land Surface Model (CLM) has proposed an alternative approach
for utilizing monthly climate outputs as forcing data since version 4.5, and
it is called the anomaly forcing CLM. However, such an approach has never
been validated for crop yield projections. In our work, we created anomaly
forcing datasets for three climate scenarios (1.5 ∘C warming, 2.0 ∘C warming, and RCP4.5) and validated crop yields against the
standard CLM forcing with the same climate scenarios using 3-hourly data. We
found that the anomaly forcing CLM could not produce crop yields identical
to the standard CLM due to the different submonthly variations, crop
yields were underestimated by 5 %–8 % across the three scenarios (1.5, 2.0 ∘C, and RCP4.5) for the global average, and
28 %–41 % of cropland showed significantly different yields. However, the
anomaly forcing CLM effectively captured the relative changes between
scenarios and over time, as well as regional crop yield variations. We
recommend that such an approach be used for qualitative analysis of crop
yields when only monthly outputs are available. Our approach can be adopted
by other land surface models to expand their capabilities for utilizing
monthly climate data.