24/01/17 – Survey in Rongxian: The Hits and Misses

DSC_0248 (2).JPG
Farmland survey – there’s a boulder with a niche around here somewhere, I know it. Hey, random kids who’ve been following us for the past hour for no reason, do you know anything about spatial analysis?

Our survey of the Buddhist rock-cut sites in Rongxian County was short and sweet, but also full of as many successes as there were bumps in the road to get there. In six-and-a-half days, we covered, plotted, and recorded quantitative and qualitative data from eleven total Buddhist niche sites. We also discovered two ancient roads, but we’re still trying to wrap our heads around what the spatial distribution of Buddhist niches across various sites along Sichuan’s Salt Road might mean, although I’ll share a hunch we’ve got below. That’ll come out as we begin to interpret the sites in relation to both published theory and the results of our mapping, but for now I thought I’d share some of the highs and lows of the entire experience.

Rongxian Google Earth Map 2.jpg
The very approximate area of our survey, with extreme emphasis towards the center of modern Rongxian.

High: Taking photos of the three big Buddhas.

Definitely the most fun part of the entire six days, I spent three mornings and an afternoon photographing a 5m, 12m, and 36m set of colossal (which is apparently anything over 4m) Buddha statues.

Gu-fo-si – a beautiful functioning monastery built surrounding this restored Tang Dynasty Buddha that ten years ago was sitting headless in a landfill by the river with its head ten meters away half-submerged by sewage.
The second-biggest Buddha at Rongxian, (known as Er-fo-si, literally second-Buddha-temple), set in the compound of a modern monastery run by nuns.
Da-fo-si (literally Big Buddha Temple) – you’ll see why the world’s second-largest Buddha gets little to no press after seeing Leshan in my next entry.

It takes a lot of patience to take the photographs needed to make a good 3D model, and while my other models have been on a much smaller scale – a niche, a wall-carving, a statue, even a cave-temple which required no less than 40 photos but no more than 120 – these three were enormous undertakings that both my and Francesca’s computers are bellyaching over. Gu-fo-si required 215 photos, Er-fo-si, so far the most promising of the three models, required 270, and Da-fo-si, which may or may not come out well because I had to take the photos surrounding the interior construct of the pagoda built into the rock surrounding the Buddha, required a whopping 380 photos.

In order to photograph a statue of this magnitude, you need to meticulously photograph every angle of every nook and cranny clearly, as well as at multiple vantage points. It took me nearly three hours to photograph all of Da-fo-si, and it also required multiple visits and some forays into places surrounding the site with both steep drops and no railings.

In the case of Er-fo-si and Gu-fo-si, the issue wasn’t the volume of vantage points but the scarcity in the statues’ enclosed areas, so sometimes I had to improvise:

The abbot lets anyone up who can climb up…so far, I’m the only person who’s succeeded without scaffolding.

Still, 3D imaging is an incredible way of reconstructing figures not only for academic interpretation away from the site, but also for purposes of heritage – many of the monuments destroyed by ISIS at Palmyra can be reconstructed using models generated from land and aerial photographs. Da-fo-si, for example, is thought to be somewhat at risk of collapsing due to poor preservation techniques in the past (as well as being on the side of a cliff prone to rockslides), and photogrammetry can potentially be used by architects and heritage staff to identify measures for best protecting sites at risk. I’m eventually giving my models to the County of Rongxian government, so they can use them as they see fit.

Low: Actually plugging in the images to Photoscan and trying to construct the models.

Waiting for Photoscan.PNG
The all-too-familiar screen of waiting half a day for your model to finish step 1 of 4….

This is a recent one, but I wondered whether it would be an issue earlier. Oftentimes, photographers and digital imagers sort the photos for their larger 3D images into chunks and allow Photoscan (or other programs) to put together those chunks once they’ve gone through the four main processes of photogrammetry (photo alignment, creation of the dense point cloud, the generation of mesh, and the creation of texture). I, being the Icarus-type go-getter I am, aimed a bit too high and tried to make all three of these 3GB+ models in single chunks. Er-fo-si crashed both mine and Francesca’s computers twice each, so it’s stuck here:

Erfosi without a waist.PNG
Er-fo-si is currently waiting in purgatory without a waist or forearms for a computer that won’t crash in the “generating mesh” mode.

This issue is definitely reparable, and I’ll probably creating chunks during the last four days I have in Beijing following our return to Peking University on the night of January 28th, but it was a bit of a buzzkill and just another thing I’ll have to play catch-up on when I get back into a world with fast desktops and no political firewalls.

High: The Physical Act of Surveying.

Fifty niches carved into a single boulder, side one.


There’s something about these type of sites that bring out both the explorer and survivalist in an archaeologist. Much of what we were surveying had been already identified by Francesca’s team under Yu Laoshi last year, but spatial analysis of these sites was new so we got to analyze the landscape. And when I mean analyze the landscape, I mean all of it. Everywhere we went, GPS in hand and laser-measuring tape in pocket, we attempted to experience the sites we were studying from all possible ancient vantage points:

For example, sitting in niches…
Or sitting on boulders with ancient handholds…
Or trying to identify the paths of ancient roads by climbing down staircases, and the fact that those staircases have begun to deteriorate is a complete understatement…
Or scrambling through brambles up a steep incline to the base of a sheer cliff face where a niche may or may not have stood, taking a GPS point, and sliding back down five feet on your butt back where you started. On your slide downwards, you notice that the niche you just mapped has a perfect vantage point to a similar-sized niche on the same valley, is located on the same ancient road, is similar-sized, and was likely constructed during the same time period. You’ve just done spatial analysis while falling on your ass – good job!

The landscape we were working with was one that went through several important phases of development, but the most telling one is that most of the activity in Rongxian between the 6th – 21st centuries had moved downward from a monastery perched high in the hills, so in order to both map and classify, we had to climb a lot, walk a lot, and scramble up a lot of terrain that made Maijishan look like a cakewalk. The niches themselves were cool too, but admittedly were eroded enough to justify our study of where they were instead of what was in them.

Low: Two days of shivering to begin the survey.

We had lunch at a random house in the countryside near one of our survey sites and all I could do to stay warm was to physically put my hands in the host’s oven as they tried to wash dishes.

Between January 19th and 20th I pretty much had hypothermia, having no ability to retain any heat and arguably no ability to sleep much more than an hour each night, and for a day and a half of those two days I was still out and surveying. Francesca says it’s caused by an overall inability to move easily from dry-cold to wet-cold climates, as was the case when we moved south from Tianshui to Chengdu relatively quickly, and that I would get over it in a few days, which I did. It wasn’t all bad, and we did manage to climb a few flights of stairs to niche-statues without much complaint on my end…

Repainting in really gaudy colors is a modern practice for ancient shrines. Some are somewhat tasteful…


While others look like someone twenty years ago said, “Okay, so we’ve got a bucket of gold paint and no one with any real artistic background among us? No problem – slap it on boys!”


…until Francesca said we were going to hike this thing because there were some niches at the top:


My repeated reaction from top to bottom: FFFFFFUUUUUUUCCCCCCCKKKKKK

That was the straw that broke the camel’s back.

Now, on a day with less fog, no heat retention, and my head feeling less like a 200 pound bowling ball, this would have been awesome, but that day, no. There was no love to be felt for this mountain. Even with a shrine at the top that had replaced their Buddhas with gold-painted images of Mao, Deng Xiaoping, Hu Jintao, and Xi Jinping, which was in itself pretty funny, I did not crack a smile.

High: The monasteries and the nuns.

We spent a considerable time in and around monasteries during our surveying – it makes sense, considering that many Buddhist sites are still worshipped today by a community with open faith today that went underground during the Maoist Period. At Er-fo-si and Gu-fo-si, two fully functional female monastic populations live, pray, and practice Buddhism with local populations of worshippers who come through on their lunch hours to pray. The monastery (not convent) is made up of women in their late fifties to mid seventies or older, who were (understandably) apprehensively supportive of our 3D modeling project and allowed us to move offerings off the altar in order to access the various corners and wall-carvings surrounding each Buddha.

They were so supportive at Er-fo-si that the moment I got dust on my hands getting up on the Buddha, they eagerly passed me a brush the moment I took a break from photographing. When the brush stopped being effective against the dirt that had accumulated for God-known-how-long on the Buddha’s shoulders, arms, and lap, they passed me a broom. And when the broom simply wasn’t getting the shine the abbot wanted on their golden Buddha, she started running wet cloths back and forth to me and got me to polish and clean it from head to toe.


The day I gave Buddha a bath…

I asked the abbot, through Francesca, when the last time they’d cleaned the Buddha like this, and she responded with a shrug and the entire monastery burst out laughing. So on our way out, she told me to come back next year so I could clean it again!

At Gu-fo-si, meanwhile, it took a bit of prodding for the nuns to let us model the Buddha, but eventually they came through. Unlike at Er-fo-si, though, where she mapped the niches in and around the monastery while I modeled the Buddha, Francesca had nothing to do. So she had her fortune told, and when I finished, so did I:


The fortune works like this: you first put 2RMB into the donation slot – that’s the most important part or else the nun will scold you for trying to get off with a freebie fortune :P. Then you shake a wooden bucket of sticks and the one that pops out first is your fortune number. Then you go to this giant book of fortunes and find your number and BOOM there it is. Mine was san-shi-san (33), and Francesca’s was liu-shi-liu (66). Weird…

My fortune outlined four main things according to the abbot (and for all my Chinese-reading friends please tell me if she’s lying): 1) you’ll travel far, very far, 2) you’ll make powerful friends and allies, 3) your health won’t be great, and 4) keep your family close. In China, fortunes apply to not just you but your family, so anyone could be the receiver of any of these fortunes. I’m hoping 3) is just an omen and not a reality, but 1), 2), and 4) I definitely can live with. I really don’t know how much more remote I can get than the Flaming Mountains in Turfan, though…

Francesca’s, meanwhile, said 1) she would get married this year and 2) she should go home and be with family. This was apparently a very good fortune, and we all had fun as Francesca mulled over the prospect of getting married before the end of the calendar year:

I’ve always found that sincerely religious people are some of the kindest and most content in the world. It’s those ones who have political aims you have to worry about…

They also have an important message for all us men out there who are somewhat apathetic to our aim:


Low: The GPS

No I did not drop off a cliff 200m I’m sitting in my hotel room!!!!!!

My worst complaint was one that gave us a headache through the entire six days. It wasn’t the GIS, it wasn’t our methodology, it wasn’t even our research questions still being formulated as we were climbing up a cliff on Day Five, it was our equipment. Through no fault of its own, my GPS was the bane of our existence.

Now, to the benefit of Garmin, this was the first survey trip I’ve taken where I’ve used my own GPS. But when you climb up a cliff and it tells you that you’re FOUR METERS LOWER THAN YOU STARTED – then you have issues.

Hmm, so I’ve been climbing for ten minutes and I’m now lower than when I started…oh yeah, that whole cloud cover thing.

Yep, that cloud cover/fog/throwing off the barometer thing. That’s what I forgot: a GPS isn’t perfect – it’s just a machine that works best when the sun is shining but has some pretty good approximations in any other condition. On the cloudiest day my GPS had an error reading of 10m in the middle of a field and the location arrow was bouncing around like a pinball, and every point was off about 8-15m from where we took it (considering we retook all those points today and they were much more accurate in the sunlight with no haze).

It made trying to map feel like pulling teeth some days, a waste of time during others, and a break-down-and-break-the-GPS type of feeling between the two. But looking at the data now, we were able to trace out where specific clusters of niches, rather than specific niches, were located, which has never been analyzed before. We also were able to identify an ancient road contemporary with most of the niches of the Song Dynasty (960 – 1279 CE)that led from Da-fo-si, up, down, and around the walls of Longdong Canyon, and back along a plateau that led to Rongxian’s main pagoda:

The pagoda also has a vantage point over Er-fo-si and the center of the medieval city (Buddhist monasteries were rarely ever built in cities before the Ming Dynasty in order to not be polluted by the city’s temptation and bad hygiene).

So, the GPS did exactly what we needed it to do: tell us we weren’t crazy thinking that spatial analysis + Chinese archaeology wasn’t a total waste of time, and helped us identify something that wasn’t there before that definitely influenced the way people saw the site and influenced the construction of monuments. But was the struggle worth the result? Well, let’s see who likes our paper(s)!

Super High: Zeng De 曾德 :

De Ge, you made this possible.

Before I sign off and restart my Gu-fo-si model, I wanted to thank Zeng De, who we call De Ge (Big Brother De). He’s been our host for the last six-and-a-half days, and while he’s really got no reason to help us (he works for Rongxian County, and while his office is at Da-fo-si, I learned recently he has no real affiliation with archaeology), he treated us much better than I’ve ever been treated by anyone while abroad. He made sure we were taken care of with meals, transportation, and often came with us to sites when he wanted to get out of the office – Francesca says he loves this stuff and treats archaeology like a hobby. He even stuck by us when a labor dispute with the people who constructed his Festival of Lights caused him to argue on the phone for hours at a time – that’s pure host-manship. Thank you.

It’ll be odd to suddenly leave Rongxian, as it’s the place we’ve stayed the longest (the second-longest being our four days in Turfan…that seems like so long ago), but I do have to start thinking about my work in Cambodia. The next couple of days will be spent going to see the world’s largest Buddha at Leshan and doing a few fun things in Chengdu (for example, see pandas and visit a museum of ornate ancient bronzes), but as of tonight the fieldwork is finished – no more niches.

I forgot to mention that we ate a lot of oranges while we were hiking in the Land of the Boulder-Shrines yesterday. One farmer let Francesca pick them herself!

Andrew Harris


One thought on “24/01/17 – Survey in Rongxian: The Hits and Misses

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s