You are an expert in visual geography and scene understanding.  
You will be shown an image from a global geolocation dataset.  
Your task is to analyze the image and identify which geographical or contextual elements are visible.  
Please carefully look at the image and mark whether each of the following categories is present ("yes" or "no").  
If the element is partially visible or ambiguous, mark "maybe".

Categories:
1. manmade — artificial structures or human-made objects (e.g., bridges, roads, fences)
2. traffic — vehicles, traffic lights, road signs, crosswalks, or road markings
3. buildings — any residential, commercial, or religious buildings
4. sign — textual or symbolic signage such as shop names, billboards, or direction signs
5. language — readable text that indicates a specific language (e.g., Chinese, English, Arabic)
6. natural — natural landscapes such as forests, grasslands, beaches, or deserts
7. plants — visible vegetation (trees, grass, flowers, bushes)
8. mountains and rivers — mountains, hills, rivers, lakes, or oceans
9. weather — visible weather conditions (e.g., snow, rain, fog, clear sky)
10. agricultural — farmlands, crops, agricultural machinery, or rural fields

Please output your results in the following JSON format:
{
  "manmade": "yes/no/maybe",
  "traffic": "yes/no/maybe",
  "buildings": "yes/no/maybe",
  "sign": "yes/no/maybe",
  "language": "yes/no/maybe",
  "natural": "yes/no/maybe",
  "plants": "yes/no/maybe",
  "mountains_and_rivers": "yes/no/maybe",
  "weather": "yes/no/maybe",
  "agricultural": "yes/no/maybe"
}