bad bad bad bovines!


Upwithclimate member Brian Machovina has just published another short article highlighting the dangers of increasing meat consumption for conservation.  The article, which is entitled “Meat consumption as a key impact on tropical nature”, was published in the journal Trends in Ecology and Evolution (TREE) as a response to a previous article by Bill Laurance et al. discussing the impacts of agriculture in general on tropical conservation.  A copy of Brian’s article is included below. Interestingly, another paper was published near-simultaneously in PNAS by Eshel et al. discussing the “Land, irrigation water, greenhouse gas, and reactive nitrogen burdens of meat, eggs, and dairy production in the United States”.  This paper reinforces Brian’s ideas and shows that “…the environmental costs per consumed calorie of dairy, poultry, pork, and eggs are mutually comparable but strikingly lower than the impacts of beef. Beef production requires 28, 11, 5, and 6 times more land, irrigation water, GHG [greenhouse gas emissions], and Nr, respectively, than the average of the other livestock categories. Preliminary analysis of three staple plant foods shows two- to sixfold lower land, GHG, and Nr requirements than those of the nonbeef animal-derived calories…”  In other words, meat is bad, but beef is the food of the devil.

Meat consumption as a key impact on tropical nature: a response to Laurance et al.
By: Brian Machovina & Kenneth J. Feeley

Laurance et al.’s review “Agricultural expansion and its impacts on tropical nature” provides a valuable summary of how agricultural is affecting the diversity of tropical terrestrial and aquatic ecosystems. However, we believe that a major factor driving the loss of tropical ecosystems and biodiversity was not given sufficient attention and deserves further discussion. Of the eight points discussed by Laurance et al. as “key challenges ahead,” no mention was made of the challenges posed by increasing per capita meat consumption. We argue that rising levels of meat consumption globally, and in developing tropical countries in particular, is one of the greatest threats to tropical ecosystems and biodiversity.

Although some agricultural expansion is driven by farmers growing crops for direct human consumption, livestock production accounts for up to 75% of all agricultural lands and 30% of Earth’s land surface, making it the single most expansive anthropogenic land use.  Of the seventeen megadiverse countries – a group of countries that collectively harbor the majority of the Earth’s species – fifteen are developing tropical countries and eleven of these have increasing rates of per capita meat consumption.

China, one of the megadiverse countries, will have a strong impact on human diet-driven ecosystem and biodiversity loss by causing rapid and extensive habitat destruction well beyond its borders.  China currently houses approximately 20% of all human beings and has a relatively-low but rapidly-rising rate of per capita meat consumption (10% of diet in 1989; 20% in 2009; on trajectory to reach 30% by 2030 with a projected 1.5 billion inhabitants). Much of China’s livestock production is fed on soy grown in the Brazilian Amazon. In Amazonia, at least 80% of all deforested lands have been converted to pasture, and much of the remaining deforested areas are dedicated to export feedcrop production. Feedcrop production is projected to grow in the Amazon, with Brazil predicted to increase soybean harvests from 60 to 95 million metric tons annually between 2010 and 2030.

A rise in meat consumption is not necessary nor is it inevitable. Increasing levels of meat consumption is connected with elevated incidences of many diseases. Diets rich in fruits, vegetables, and plant-based protein sources are healthier than those containing a higher proportion of meat and dairy products.  Eliminating livestock and growing crops only for direct human consumption could increase the amount of calories that can be produced on extant agricultural lands by an estimated 70%.  This could feed an additional 4 billion people – significantly more than the projected global population growth of 2–3 billion. Much of the future population growth will occur in developing countries where low-cost, locally-available and environmentally-sensitive practices and technologies can improve production of plant-based food sources and provide necessary caloric, protein, and nutrient levels.

Based on a balance between the need to increase nutritional health and availability of calories with the need to decrease the land demands and ecological footprint of agriculture, we argue for a goal of significantly reducing the contribution of animal products in the human diet, ideally to a global average of 10% or less (this is roughly equivalent to limiting daily consumption of meat to a portion that is approximately the size of a deck of playing cards or smaller). Within the context of decreasing total meat consumption,  the spatial and climate change footprint of agriculture  can be further reduced by the preferential use of meat sources with higher energy conversion efficiencies (i.e. chickens > pigs > ruminants) and a switch to more-efficient production methods.

Reaching the proposed goal will require significant decreases in per capita meat consumption by developed countries and little or no increase in developing countries.  For example, animal products currently comprise approximately 48% of the average diet in the USA. Developing countries will need to resist emulating the animal-product rich diets of developed countries and stabilize meat consumption near their current levels. Reducing the total per capita consumption of meat and increasing the proportion of meat that is derived from more-efficient sources will enable developing tropical countries to feed more people on less land even if total caloric and protein intake increase, hence maintaining human wellbeing and reducing threats to biodiversity. Without a global per capita decrease in meat consumption, the successful conservation of Earth’s remaining tropical ecosystems, and the great biodiversity that they contain, is almost certain to fail.

[Literature citations have been removed to improve clarity but are available upon request]

Machovina, B. & Feeley, K.J. Meat consumption as a key impact on tropical nature: a response to Laurance et al. Trends in Ecology & Evolution, 29, 430-431. Available online

photos from latest trip to field

I just returned from a quick tour of our field sites in the Kosnipata Valley of Peru (Tres Cruces @ 3700m, Wayquecha @ 3000m, San Pedro @ 1500m, and Villa Carmen @ 700m).  I have uploaded a bunch of the pictures from the trip into an album in flickr which is available for viewing HERE.  The purpose of this trip was to show the field sites and our work to climate change photographer Gary Braasch and his partner Joan Rothlein.  I hope/plan to soon write future blog posts about the value of working with photographers and journalists to help reach a broader audience.

P1090091 P1090281P1090247P1090231P1090211P1090057

The dangers of carbon-centric conservation for biodiversity

REDD (reduced Emissions from Deforestation and Degradation) has become the focus of many conservation efforts and is eating up lots of conservation money.  While REDD sounds good on paper, has many serious problems, most of which have been discussed at length elsewhere (e.g., HERE, HERE, HERE).  In collaboration with Alvaro Duque and colleagues, we have just published a new study highlighting one underappreciated problem with REDD and other carbon centric conservation schemes.  Basically, by adding value to high-biomass areas (such as lowland tropical forests, deforestation and degradation may be pushed to lower-biomass areas such as highland forests (i.e., leakage).  As we show in our analyses these low-biomass forests often contain high amounts of diversity and super high amounts of endemic diversity.  Importantly a lot of the diversity in these low-biomass forests is in life forms other than trees (e.g., fern, herbs, lianas, epiphytes…).  So even in schemes such as REDD+ where biodiversity is taken into consideration, low biomass forests may still be at risk since measures of biodiversity are usually based only on trees or other large charismatic species. The end result is that while carbon-centric approaches to conservation can potentially promote the protection of some habitats and thereby reduce net carbon emissions, they can potentially have the perverse effect of promoting deforestation in other habitats and thereby actually increase overall species extinction rates. The abstract of our paper, entitled “The dangers of carbon-centric conservation for biodiversity: a case study in the Andes,” is reproduced below and the original article is available through the journal Tropical Conservation Science HERE.


The dangers of carbon-centric conservation for biodiversity: a case study in the Andes
Alvaro Duque, Kenneth J. Feeley, Edersson Cabrera, Ricardo Callejas and Alvaro Idarraga

Carbon-centric conservation strategies such as the United Nation’s program to Reduce CO2 Emissions from Deforestation and Degradation (REDD+), are expected to simultaneously reduce net global CO2 emissions and mitigate species extinctions in regions with high endemism and diversity, such as the Tropical Andes Biodiversity Hotspot. Using data from the northern Andes, we show, however, that carbon-focused conservation strategies may potentially lead to increased risks of species extinctions if there is displacement (i.e., “leakage”) of land-use changes from forests with large aboveground biomass stocks but relatively poor species richness and low levels of endemism, to forests with lower biomass stocks but higher species richness and endemism, as are found in the Andean highlands (especially low-biomass non-tree growth forms such as herbs and epiphytes that are often overlooked in biological inventories). We conclude that despite the considerable potential benefits of REDD+ and other carbon-centric conservation strategies, there is still a need to develop mechanisms to safeguard against possible negative effects on biodiversity in situations where carbon stocks do not covary positively with species diversity and endemism.

Antioquia Colombia


Species Migrating Upslope Due to Climate Change in Tropical Montane Cloud Forests of Peru May Meet a Grass Ceiling

A recent piece from the Huffington Post discussing the work of UpWithClimate team member, Evan Rehm, is reproduced below.  The original article by Keith Peterman is available HERE.


recent letter by Evan Rehm in the Proceedings of the National Academy of Sciences (PNAS) respectfully challenges prior “spurious conclusions” by authors ofan earlier article hypothesizing that “tropical montane species are responding more strongly to climate change than temperate-zone species.” This scientific debate is not about climate change — scientists are in broad agreement that climate change is occurring. And, the debate is not about species redistribution — the IPCC foundwith “very high confidence” that species redistribution due to climate change is occurring on all continents and most oceans. The issue here is more subtle, but one of great importance for species survival.

Last August, I sat down with Evan at a Starbucks in York, Pa. to discuss his tropical research. Unshaven and a bit red-eyed, he had just returned from a two-year stint high in an Andean cloud forest of Peru. He was on a brief stop-over in York to visit family before return to his academic home in the Department of Biological Sciences at Florida International University.


Evan’s research interests broadly include species adaptations to climate change and environmental ecology. He conducts his primary research in the tropical montane cloud forests of Manύ National Park on the eastern slope of the Andes Mountains in Peru. Cloud forests are characterized by the presence of clouds or persistent mist — even in the dry season — and much precipitation comes in the form of “canopy drip” resulting from the condensation of fog or mist on tree leaves. A scant 1 percent of global woodlands are cloud forests, yet they are among the most biodiverse regions on Earth.

Working at the high-elevation treeline where closed-canopy cloud forest forests meets open alpine vegetation, Evan is trying to understand how these forests will adapt to climate change.


He explains how species are shifting ranges (migrating to new areas) in order to track the temperatures to which they are adapted — some species migrate toward the poles while others shift upslope to seek cool-temperature refuges.

He says, “The distances species have to travel based on temperature change are staggering.” Recent studies show that the rate of travel correlates closely with the rate of temperature change in a given region. Evan says, “It’s not just the temperature change. Man-made barriers such as deforestation, roads, and urban areas increase the distance [and therefore] decrease the speed at which species can migrate.” A good example is lowland Amazon rainforests which are especially vulnerable to climate change.

At higher elevation, “The treeline should be one of the first and most obvious shifts in ecological edges.” However, “Trees shift much slower than mobile organisms such as vertebrates.”

Evan states, “Species ability and speed of migration is particularly important in tropical montane cloud forests. Because biodiversity is extremely high, most species occur in narrow temperature ranges.” These narrow-niche species should react quickly to climate change by shifting upslope. However, if the treeline does not shift upslope with species occurring below the treeline, then the treeline may act as a barrier to upslope migrations of other species. “This creates elevated extinction risks in Andean tropical montane cloud forests.”

Fittingly, an article titled “Will Climate Change Imperil Your Cup of Starbucks?” appeared in National Geographic just three months after Evan and I shared our cup of Joe at the Starbucks in York. The author of this article chronicled his hike through the dense Andean cloud forest with Evan’s academic advisor Ken Feeley. Feeley stated, “There are known accounts that coffee as an agricultural product is moving up the slope with farmers planting it higher and higher. The cultivated coffee-growing areas in the lowlands are seeing decreased yields.” Of course, farmer’s planting at higher elevations is not the same as natural migration of trees. Feeley predicts significant species population reductions and extinctions due to climate change in the next 50 to 75 years based on his tree-migration studies in both Peru and Costa Rica.

During my own academic research visits to Costa Rica, I’ve encountered another climate change stress on coffee called “flora loca” (crazy flowers). I first observed flora loca in March 2010 while passing through small coffee plantations just beneath the Monteverde cloud forest. The scene of flower-covered coffee trees simply did not fit the season. For a northerner like me, it was something akin to a spray of fresh daffodils in frozen December soil. Diego Calderón, an agricultural management engineer, told me that these trees were actually on their third flora loca of the season, and that due to changing weather conditions in recent years, this mistimed blooming could be linked to climate change.

Getting back to the Pervian Andes cloud forest, Evan explains why the tropical montane treeline shift may not keep pace with climate change. Although “animals have the ability to move to new areas when their current location becomes climatically unsuitable… upslope shift of plants largely depends on seed dispersal” over multiple generations. Movement of the treeline may be further limited by narrow dispersal of seeds at the forest edge, low germination rate of seeds due to the harsh alpine microclimate, competition with established grasses, and even higher intensity UV solar radiation.


In spite of the scientific debate concerning whether tropical montane species are or are not responding more strongly to climate change than temperate-zone species, one point is clear. The current upslope migration of tropical species due to climate change will encounter an ecological edge at the treeline which will challenge their very survival. Species will confront A Grass Ceiling — the alpine zone above the tropical montane cloud forest where they no longer have the ability to adapt.

By Keith Peterman

Corridors are for connecting

A recent paper in Nature Climate Change by Jantz et al. mapped out potential corridor routes that can be used to connect tropical parks while simultaneously protecting large amounts of carbon.  We (Feeley and Rehm) wrote a response this article highlighting some of the potential problems with this approach (this was subsequently followed by a reply by the Jantz et al.).  The discussion about how to prioritize conservation corridors was also featured in a recent National Geographic News Story available HERE. Our response published in Nature Climate Change is included below:

Priorities for conservation corridors
by Kenneth J. Feeley & Evan M. Rehm


To the Editor —

image from

The motivation for establishing corridors should be first and foremost to facilitate the moment of individuals and species between otherwise disconnected habitats. image from

Jantz et al. take advantage of new, high-resolution estimates of biomass and vegetation carbon storage (VCS) to map areas throughout the tropics that, if protected, could simultaneously connect existing protected areas while also retaining large carbon stores. This study highlights how the growing wealth of remotely-sensed data can be used to intelligently and purposely design protected areas. Given the recent emphasis on carbon sequestration in establishing and funding protected areas, it is understandable that the authors took a largely carbon-centric approach when identifying their proposed conservation corridors. We argue, however, that there are more important factors that should be considered when evaluating and prioritizing potential corridors.

The principle motivation for establishing corridors is not to protect VCS but to allow individuals and even entire species to move between otherwise disconnected habitats. Corridors should ideally be set up to connect similar habitats and cross through habitats similar to those being connected. Jantzet al. did not consider the habitat type or the species composition of the areas that they were connecting. Likewise, they did not consider the type of habitats contained within the proposed corridors in relation to the connected protected areas. Instead, the authors proposed corridors that would contain the greatest possible density of carbon and the greatest possible diversity of mammal species. Following these guidelines, high-priority corridors could theoretically be placed through high-biomass, high-diversity areas to connect different low-biomass habitats with distinct species compositions (for example, a corridor of rainforest connecting a savannah park to a dry forest park). In several places, such as in the southeastern Amazon, Jantz et al. suggest corridors through areas that are already heavily-modified and under intense human cultivation.

When prioritizing potential corridors for conservation, it is also important to consider climate-driven species migrations. Climate-driven species migrations are different from the more traditional movements of individuals and species in that they are directional, with species migrating from climatically unsuitable areas to more suitable ones. For example, warming in the tropics will drive species migrations from the lowlands to the colder highlands. By combining species distribution models with general circulation models, it is possible to predict where species are now and where they will need to be in the future, thereby helping to guide where conservation corridors should be established.

Even accepting a carbon-centric viewpoint, Jantz et al. have probably overestimated the long-term VCS in their proposed corridors. By definition, habitat corridors are long and skinny (on average, the proposed corridors are 41–55 km long and 2–3 km wide) and thus a large fraction of the total corridor area will suffer from edge effects. These edge effects can include, for example, biomass/carbon collapse due to the increased mortality of large trees at distances of up to 100 metres from the forest edge and increased susceptibility to fire at distances of up to several kilometres from the edge. The habitat within corridors will inevitably degrade due to pervasive edge effects, causing VCS to decrease over time. In contrast, protecting large, contiguous blocks of natural habitat will result in more stable carbon dynamics as a larger proportion of the protected areas will be core habitat. To protect biodiversity in a changing world, we need an extensive network of large, well-connected protected areas. The corridors that allow for these connections should be designed with species movements, not carbon storage, as the priority.

Feeley, KJ & Rehm, EM. 2014. Priorities for conservation corridors. Nature Climate Change. 4(6): 405-406.