Daily Land Surface Temperature Reconstruction in Landsat Cross-Track Areas Using Deep Ensemble Learning With Uncertainty Quantification

Shengjie Kris Liu: skrisliu@gmail.com

S. Liu, S. Wang and L. Zhang, 2025. "Daily Land Surface Temperature Reconstruction in Landsat Cross-Track Areas Using Deep Ensemble Learning With Uncertainty Quantification," in IEEE Transactions on Geoscience and Remote Sensing, doi: 10.1109/TGRS.2025.3643985.

This is a GitHub repo at github.com/skrisliu/delag [code & data].

Minimum Run [NYC]

bash - Minimum Run [NYC]: Task #1 Clear-Sky With Real-World Cloud Patterns 
$ git clone https://github.com/skrisliu/delag.git
$ cd delag
$ curl -L https://github.com/skrisliu/delag/releases/latest/download/nyc.7z -o nyc.7z
$ 7z x ./nyc.7z -o./nyc
$ python nyc_task1_01.py
$ python nyc_task1_02.py

Data Download

NYC

  1. NYC Dataset: nyc.7z
  2. NYC Predictions LST 2023: nyc_pred_2023.zip

London (LDN)

  1. LDN Dataset: ldn.7z
  2. LDN Predictions LST 2023: ldn_pred_2023.zip

Hong Kong (HKG)

  1. HKG Dataset: hkg.7z
  2. HKG Predictions LST 2023: hkg_pred_2023.zip

NYC Sample Data

The generated temperature data can be downloaded at [NYC 30 m Daily LST 2023]. NYC Sample Data

Minimum Run [NYC] Task #3: Estimate near-surface air temperature as indirect validation (Fig. 12)

bash - Minimum Run [NYC] Task #3: Indirect Validation via Estimating Near-Surface Air Temperature 
# curl -L https://github.com/skrisliu/delag/releases/latest/download/nyc.7z -o nyc.7z
# 7z x ./nyc.7z -o./nyc
$ curl -L https://github.com/skrisliu/delag/releases/latest/download/nyc_pred_2023.zip -o nyc_pred_2023.zip
$ 7z x nyc_pred_2023.zip
$ python nyc_task3_plot_airtemp.py