RS-PestDyn

The Remote Sensing insect Pest Dynamic model is an insect pest population dynamics model driven by temperature, which is derived from satellites at a daily basis. The satellite data used is the 1km land surface temperature (LST) product of MODIS, which is directly downloaded from Google Earth Engine. LST is then used in RS-PestDyn as a daily input to the model while a series of ordinary differential equations (ODEs) are solved simulataneously through the Runge-Kutta method for daily simulations of insect pest populations. Currently, RS-PestDyn simulates the population of a generic pest for a single life stage and for one day only. It still lacks several processes, which are matained constant at this stage. For more information on the basics of this model and a previous version of the model (one-dimensional model using Euler solution) see Blum et al. 2018.

Main Program:

• 'run_RK4_2D.py'

Called Procedures:

• 'Download_Temp_GEE.py' - download TIF files of LST from MODIS as a batch from GEE
• 'RK4_2D_procedure.py' - solves the ODE using Runge-Kutta
• 'display_plots.py' - displays results (as a 2D matrix of % population)

Still Needs To:

1. extend running period to a seasonal timeframe (currently for 1 day)
2. extend the model for various life stages (currently only 1 life stage)
3. improve numerical solution accounting for error expansion through time.
4. add LC/LU classification to account for plant host preference
5. add plant growth dynamics (host stage)
6. add migration between grid cells