Reply to: Fire activity as measured by burned area reveals weak effects of ENSO in China
Bibliographic record
Abstract
The different wildfire data used in Fang et al. (2021) and Resco de Dios et al. may be the primary reason for the discrepancies between the two studies. We developed the Wildfire Atlas of China (WFAC), a robust ground-truthed fire occurrence dataset, in which fire occurrences were detected by multiple satellites (including MODIS) and carefully validated through surface observations by local forestry departments 1 . Even though MODIS-derived fire products, such as active fires and area burned, have much improved in recent years 2 , biases still remain in these products because of surface topography, fast vegetation growth rates, and clouds 2 . In addition to these biases, crop data gaps due to clouds occur in these fire products 2 and this is especially the case in cloudy subtropical China. As a result, fire patterns in southwestern China have been found to be inconsistent between MODIS fire products and ground observations 3 . To address these biases, we calculated fire occurrences from MODIS fire points 4 and further distinguished between wildfire and crop fire in the fire occurrence dataset. We find a high fraction of crop fires in the MODIS fire data in northern China, particularly in the North and Northeast China Plains (Fig. 1 ). These northern Plains are mostly covered by agricultural land rather than forests and routine agricultural crop straw burning accounts for most of the MODIS observed fires in these agricultural areas 5 , 6 . Therefore, most fires in northern China detected in the MODIS-derived area burned dataset used by Resco de Dios et al. are crop fires, leading to an overestimation of wildfire activity outside of subtropical China. In addition to this, Resco de Dios et al. calculated the area burned over only the area south of ~35 ○ N in eastern China, which is only a fraction of subtropical China. Their suggestion that subtropical fires only account for a low fraction of fires in China is therefore geographically biased. Fig. 1: Fire occurrence comparisons between fire datasets. The fire occurrence numbers are from the Wildfire Atlas of China (WFAC) and the Moderate Resolution Imaging Spectroradiometer (MODIS) data, including MOD14A1 provided by Terra and MYD14A1 provided by Aqua 4 . a Total fire occurrences from MODIS, b crop fire occurrences from MODIS, c the ratio between total fire occurrences from MODIS and WFAC, and d the ratio between non-crop fire occurrences from MODIS and the WFAC. The fire occurrences were aggregated via combining the spatiotemporal adjacent fire points from MODIS Terra and Aqua from 2005 to 2018 into one fire event. Crop fire was picked out based on the land cover data of China in 2018 11 . The The 2 × 2 degree fire data is a sum of the fire occurrences. Full size image
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.014 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.038 | 0.028 |
| Insufficient payload (model declined to judge) | 0.004 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".