Weather Text Analysis


Climate change, a current hot political issue, is a process in which the planet has cycled through regularly in throughout time. Evidence of such changes can be seen in sediment cores, tree rings, and historical documentation. This last source of evidence, historical documentation, can be in many forms, including old forecasts, Farmer’s Almanacs, and in this case, travel journals. Marco Polo, now a house-hold name, is the story of a man that few homes truly know. His travels during the late 13th century remain a debate in historical legitimacy based on omissions of various characteristics of places he visited, exaggerated and false claims, and lack of documentation supporting his visit has led many scholars to question the course of his travels (Jackson 82).

These alleged travels took place during a major climatic transition in Earth’s history, the transition from the Medieval Warm Period to the Little Ice Age, and while little can be said from the day to day accounts, the 24 year period of his travels could present additional documentation of these climatic changes.

Using The Travels of Marco Polo, text analysis was used to investigate the weather conditions the travelers encountered on their voyages as a way to speak to legitimacy of this work and explore the transition from Medieval Warm Period to the Little Ice Age.

Marco Polo

The household name and childhood game, Marco Polo doesn’t even tell half the story of the individual holding this name. Assuming the acceptance of Marco Polo’s existence, another historically debated issue, the man was a Venetian traveler who voyaged from Europe throughout Asia and is credited as the first European to travel to China and back (Jackson 83 and Meicun and Zhang 39). Marco Polo traveled with his father and uncle  eastward from Europe, and landing in China in 1271, where the Mongol ruler Kubilai Khan used the youngest Polo as an emissary, sending him all throughout the empire for the next 17 years (Wood 1). After being granted permission to return home in 1291, the three men were tasked with delivering a Princess for her marriage in Persia, however, the difficult sea voyage left only 18 of the 600 original men alive, and from there travel consisted of horseback and sea to Venice (Wood 1). The book, most likely produced in 1298, was written by Pisan author Rusticello during their internment in a Genoese prison (Jackson 84). Despite these accounts, there is much debate surrounding the legitimacy of the Polo’s travels.

The travels of Marco Polo have been a long-standing debate in historical literature, with scholars citing lack of detail, lack of corroborative sources, and falsified and exaggerated observations. Many scholars assert that omission of characteristics about that Mongol Empire, The Great Wall of China, feet binding, and tea are suspicious (Jackson 82). There are also only a handful of sources that support his presence in these foreign locations and is not mentioned in any Chinese historical documents (Jackson 82). And finally, Polo claims that he and his accompanying family members participated in the seize of a Chinese city, however, this event is confirmed to have happened one year prior to their arrival (Jackson 82). While some of these features are more then incriminating, Polo’s travel account may aid in his defense. As he acted as an emissary for the Khan, he spent much of his time away from the empire which could explain the omission of characteristics such as tea, and as for the Great Wall of China, it’s difficult to confirm the building progress, and while walls were certainly in the empire at this time, the exact magnitude is unknown (Jackson 83).

Climate During the 13th-14th Centuries

Marco Polo’s travels during the transition from the Medieval Warm Period to the Little Ice Age can be utilized a source of evidence in the Marco Polo debate and provide first-hand accounts of a climatic shift taking place. The first of these climatic events, the Medieval Warm Period, lasted from roughly 900-1300 AD and represents a period in which temperatures reached levels comparable to those of the late 20th century (Mann “Medieval” 1). During this period Europe experienced overall mild climate conditions, that allowed for increased agricultural yields and the growth of subtropical flora, such as olive trees, to grow in regions of the continent (Mann “Medieval” 1). At this time mountain glaciers receded, and sea surface temperatures were warmer then past periods (Mann “Medieval” 1). However, this period would not last indefinitely, as the ‘Little Ice Age’ would soon shuffle in. Following this warmer period was the Little Age, which was a time of glacial expansion and is commonly tied to frequent crop failures (Mann “Little” 1). While scholars have used The Travels of Marco Polo, to aid in better understanding global climate change history, this project aims to use the text as a way to better contextualize the climatic shift occurring and add new interpretations to the Marco Polo debate (Klimenko 370-371).

As these two events are well document in the geological and historical records, they be used a way in which to interpret the climatic conditions Marco Polo experienced throughout his travels, and while climatic conditions are subject to various atmospheric conditions, they can be used as another line of evidence in which to interpret the validity of the travel accounts.



The document used in this text analysis is The Travels of Marco Polo. The 17th century Henry Yule translation was obtained from the Guttenberg Project’s website. As the text is a translation of the original work, we acknowledge the limitations of working with these sorts of materials. These sorts of shortcomings may come in the form of translations errors, misrepresented word contexts, and biases. The Henry Yule translation was selected for a number of different reasons including; ease of access, translation date, and text format. Only the actual text of the book was used through this analysis. Notes and prefaces were excluded from the study. A total of 24 words were searched for throughout the text and analyzed in various fashions using Voyant.


The text of Marco Polo was copied over to Voyant for analysis. At this time a list of words was constructed that relate to weather conditions. The accompanying list only includes those words which were present in the text, for example common weather words such as snow and ice were not present in the text and were thus omitted from analysis. While these words are generalizations of weather events, they provide a clue in which to explore the sorts of conditions Marco Polo encountered on his travels and offer points of interests within the text. Voyant offers a wide range of tools that allow for users to customize their investigations to their research questions. In this project a total of five different tools were used in this weather text mining analysis.

Words selected for analysis:

  1. Weather
  2. Climate
  3. Conditions
  4. Temperature
  5. Cloud
  6. Rain
  7. Sun
  8. Earth
  9. Cool
  10. Cold
  11. Warm
  12. Warmth
  13. Thunder
  14. Thunderstorm
  15. Lighting
  16. Storm
  17. Moon
  18. Snow
  19. Hot
  20. Season(s)
  21. Summer
  22. Winter
  23. Air
  24. Fall

Trends and Bubblelines

First Trends were used to track how often each of the words were used throughout different portions of text in the form of a graph. For Trends the word ‘Earth’ was omitted as it was a major outlier and skewed the visualization. While the Trends feature does not show the contextual relationships between the words or context surrounding their use, it is a useful tool allowing researchers to take a quick glimpse at where these words were used in the text in comparison to one another.

Another useful tool that displays the frequency of words is Bubblines, however, in addition to frequency, it also displays the words repetitions throughout the text in a clearer fashion then trends. Despite this increase in clarity, Bubblelines still falls victim to the same limitation as Trends, its inability to display contextual details of the words use in the text. Regardless, Bubblines is useful in looking at the repetition of words use throughout the text and providing a quick glimpse at the relationship between these terms.

In this research, the trends and Bubblelines tools has been most useful in looking at the different sorts of weather events encountered by the Polos during their travels and their chronological relations to one another.

I had expected to see clear correlations between related words such as summer and hot, however, that was not the case in many situations.

For example, in the first section of the text, summer and snow appear. While doing the contextual research, I also hoped to see some evidence of climatic shifts, however, instead words relating to temperature, such as hot and cold, appear to be more common in the first half of the text, whereas terms relating to specific weather events, rain and storm, are more common in the latter half of text. Also, interesting to note is that snow and winter are only mentioned in the first half of the text, whereas terms such a sun and wind are mentioned throughout the text. Only two words only appear at the end of the text; moon and thunder. Together, both tools suggest that Marco Polo encountered winter conditions more often in his early travels, while experience warmer and summer-like weather throughout, and more inclement weather events such as storms towards the later part of his journey.


Links analysis using Voyant

Links is another useful tool in Voyant that allows that allows for researchers to see connections between words of interest and their collocates. This allows for researchers to gain a better understanding of the terms and frequently associated words. Particularly in this project, Links has been useful in determining which sort of weather related terms are used in association with one another and the links between these events. This tool also provides slightly more context then the previous two tools, allowing users to better understand the ways in which these terms were used within the text. For example, when looking at the word storm, India is listed a collocate, which allows for researcher to interpret this a possible connections with storms in India, however, it is interesting to note that Marco Polo never stepped foot in India during his travels.

This in itself is interesting, as it suggests long distance connections talking about this weather situation, but also can start to call into question the legitimacy of the story.

In this project Links has played an important role in showing the connections between weather related words throughout the text while providing more context for interpretations. While the possibility of storms in India is interesting, many other interesting connections are exposed. One of the more interesting is the that the link between weather and changes. This leads to the interpretation that Marco Polo most commonly recounted times in which the weather was changing. Also, of interesting note is not only does it seemed as though Marco encounter storms during his time in India, but also in Buadas, modern day Baghdad . And as seen in Bubblelines, two out of three of the times storms is mentioned in the first half of the text, also accompanied with occurrences of summer and rain.

A final note relating to the use of Links in the text is not as much related to the experience weather conditions, but more related to the ways in which weather was discussed. For example, while there are many accounts of terms being used to described temperatures or precipitation, there are also many accounts linking the climate to agriculture and supernatural entities. For example, air links to gods, and summer to abundant, and while not explaining any exact weather experience, it does shed light on the ways in which these conditions where discussed. In this case, understandably so, weather was linked to agricultural output, and also to higher powers.

Principle Component Analysis

Principle Component Analysis
Principle component analysis of weather related terms in The Travels of Marco Polo

Principle Component analysis is another tool available to Voyant users through scatterplots that allows researchers to explore data through statistical means. Principle Component Analysis “is a dimension-reduction tool that can be used to reduce a large set of variables to a small set that still contains most of the information in the large set” (Thorne 2). In this project Principle Component Analysis (PCA) has been most useful in looking at relationships between the frequency of weather related terms.

In this project, Principle Component Analysis was used on the 24 words of interest in the text to test their relationships between one another. Of the resulting statistical analysis, four groups were formed.

Group One (Blue Group):

  1. Fall
  2. Rain
  3. Winter
  4. Condition
  5. Warm
  6. Thunder
  7. Lighting
  8. Cool

Group Two (Pink Group):

  1. Earth

Group Three (Green Group):

  1. Snow
  2. Air
  3. Season
  4. Weather
  5. Cold
  6. Moon

Group 4 (Purple Group):

  1. Summer
  2. Storm
  3. Hot
  4. Sun

Looking at these groups a few interesting characteristics stand out. For example, group four is particularly interesting as it solely contains words pertaining to summer. Also interesting is group three, which contains only Earth. This is interesting because in trends Earth was excluded as an outlier, and once again remains an outlier. As the aim of principle analysis is to show possible correlations between these words, it is most useful in analyzing similarities in the words usage throughout the text.


The final tool used in this project was Contexts. This tool allows researchers to look at each occurrence of a word throughout the text and a bit of the surround text in order to provide users with more contextual information around the terms use. For this research, Contexts has been useful in understanding more context around the appearance of these terms.

Results and Implications

Weather Links

Looking at the results of each individual tool, the weather events experienced by Marco Polo loosely fit with climatic changes taking place at the turn of the 13th century and describes the ways in which weather was discussed historically and it’s role in day to day life.

Using Links, weather is associated with the term changes, leading to the interpretation of changes in weather throughout his travels, a change we see in overall climate during this time period with the impending Little Ice Age.

Storms and India Link
Stroms links

From Links we also learned that while in India Marco Polo experienced storms, so using this information in combination with Bubblelines, we see that these storms were also mentioned simultaneously with wind, rain, and summer. While little can be said other than that these weather conditions were experienced around the same time, it does show the Marco Polo experienced summer like weather early in his travels, and loosely correlates with the end of the Medieval Warm Period. While the Medieval Warm Period doesn’t necessarily mean that winter never happened, it is coincidental that terms pertaining winter appear solely in the first two thirds of the text. While not necessarily fitting with the climatic shift occurring during this period, it does follow the trend that general weather terms appear more commonly in the first part of the book. This can be explained by fact that on their travels home, specific weather-related terms were used more commonly to describe the rough travels home.

Contexts and Links were also useful in looking at the ways in which weather was discussed historically. As agriculture was both a key source of subsistence and work, it is unsurprising that such terms have with various words such as abundance, horses, and other godly entities. Both associations show the emphasis placed and agriculture and unearthly entities when discussing weather.

Agriculture Link
Links related to agricultural production


Using The Travels of Marco Polo, text analysis was used to investigate weather conditions the travelers encountered on voyages as a way to speak to legitimacy of this work and explore the transition from Medieval Warm Period to the Little Ice Age from first-hand accounts.

Climate change is typically a slow enough process that makes surface level observations during this period difficult, and due to the limited sample of this project, little can be said in respect to the transition from the Medieval Warm Period to the Little Ice Age, however, we do learn about the different ways in which weather was discussed and the sort of roles it played throughout Marco Polo’s travels. Increasing the sample to include other documents of this era and other’s mentioning Marco Polo would aid in not only addressing weather transitions during this period, but also continue to address the debate surrounding Marco Polo.

Created By: Amy H.