 Trying to figure out what to wear tomorrow based on the weather? You'll probably check a weather forecasting app, right? Now pretend you're a farmer, and you're trying to figure out what crop will be the most profitable to plant next year based on the weather. What about 50 years from now? Instead of checking a short-term weather forecast for these answers, scientists are working to advance the field of ecological forecasting. Ecological forecasting is similar to weather forecasting, but instead of only predicting weather a couple days out, ecological forecasting attempts to predict environmental conditions based on interactions between organisms and their environment on a longer time scale. Ecological forecasting is really important because it can provide a glimpse into the future state of an ecosystem, from wildlands, like forests and marshes, to managed lands, like farms and cities. This sort of forecasting will help landowners and policymakers anticipate upcoming environmental change and possibly mitigate the impacts. To be able to forecast how an ecosystem will be in the future, scientists use data collected from the past and present in combination with models. No, not those kind of models. Ecological forecasting models are a series of equations produced by scientists that are fed data about air, land and water. Organizations like NEON provide these data freely so that scientists can produce forecasts on how an ecosystem will be in the future. But how can we know if the forecast these models produce will be accurate? On a short time scale, scientists can build a forecasting model that will make a prediction about the near future, say four months from now. After the four months have passed, they can compare the predictions to what actually happened and use those observations to improve the model. Each iteration allows scientists to peek further into the future of that ecosystem. This is similar to how the National Weather Service began its forecasting process in the 50s. Back then, scientists were predicting the 36-hour forecast with about 23% accuracy. The more weather forecasts were produced and compared with actual data, the weather service has been able to adjust and improve its 36-hour forecasting model to almost 85% accurate. One useful application for ecological forecasting could be to predict corn production in Iowa 10 years from now. Another might be forecasting the types of disease-carrying mosquitoes that will live in New York in 50 years. Waiting that long to compare predictions to the results just isn't feasible for advancing this science. An alternative is to build a model based on a historical time period, like using data collected from 1960 to build a model about what would happen 50 years later. And since we already have that data from 2010, we can test it. These types of comparisons allow scientists to continually improve the accuracy of their models and reduce the amount of uncertainty in them. Uncertainties in the models are the variables that scientists can't fully account for, but that's not to say they can't eventually be with new technology, more data, and discoveries. Uncertainties are an essential part of ecological forecasting because they help to define boundaries on the forecasted state of the ecosystem. The more uncertainty in the model, the less scientists can precisely project the future state of the ecosystem. As the model projection extends into the future, the uncertainty around the forecast also increases. But like the weather forecasting models have shown, with each forecast, test, and update, these decrease and accuracy improves. With that, we know more about the range of scenarios that will not happen and the scenarios that are likely to happen become easier to identify. As we move toward the future and ecological forecasting becomes more accurate, scientists can better address processes, like making forecasts for how an ecosystem will react if we institute a policy to reduce greenhouse emissions by 25% by the year 2040. So, pulling it all together, ecological forecasting is a really useful way to gain a better understanding of our ecosystems and how they are changing. It's a way scientists are helping policymakers and landowners take a peek into the future to anticipate major environmental change. Like the weather service models have shown, ecological forecasting accuracy will improve through time as more predictions are made, tested against actual data, and then refined. As a result, ecological forecasts will inform planning efforts that allow humans and the environment to thrive moving forward.